From 737e1334674b1b6752fdba42c8cec5fbb2db1aca Mon Sep 17 00:00:00 2001 From: Ali Mohammad Date: Tue, 10 Nov 2026 10:29:11 +0300 Subject: [PATCH] chore: point at the renamed SDK repo LinuxDevil/lousho Co-Authored-By: Claude Opus 5.5 --- .github/workflows/sync-sdk-docs.yml | 2 +- README.md | 2 +- acp.mdx | 2 +- agent-directories.mdx | 2 +- agent-forge.mdx | 4 +- agent-spec.mdx | 14 +-- ai-sdk-ui.mdx | 2 +- api-overview.mdx | 2 +- approvals.mdx | 2 +- ar/acp.mdx | 2 +- ar/agent-directories.mdx | 2 +- ar/agent-forge.mdx | 4 +- ar/agent-spec.mdx | 14 +-- ar/ai-sdk-ui.mdx | 2 +- ar/api-overview.mdx | 2 +- ar/approvals.mdx | 2 +- ar/auth.mdx | 2 +- ar/build-a-coding-agent.mdx | 2 +- ar/channels.mdx | 2 +- ar/claude-projects.mdx | 2 +- ar/cli.mdx | 2 +- ar/cloudflare-workers.mdx | 2 +- ar/code-mode.mdx | 2 +- ar/coding-agents.mdx | 2 +- ar/compaction.mdx | 2 +- ar/comparison.mdx | 4 +- ar/configuration.mdx | 2 +- ar/connectors.mdx | 2 +- ar/decisions.mdx | 4 +- ar/deployment.mdx | 2 +- ar/durable-execution.mdx | 2 +- ar/errors.mdx | 2 +- ar/evals.mdx | 2 +- ar/examples.mdx | 34 +++--- ar/examples/coding-agent-workflows.mdx | 2 +- ar/examples/connectors.mdx | 2 +- ar/examples/heal-benchmark.mdx | 2 +- ar/examples/hostinger-deploy.mdx | 4 +- ar/examples/phase-pipeline.mdx | 2 +- ar/examples/waves.mdx | 4 +- ar/executor-api.mdx | 2 +- ar/flows.mdx | 2 +- ar/guardrails.mdx | 4 +- ar/handoffs.mdx | 2 +- ar/hooks.mdx | 2 +- ar/hosted-tools.mdx | 2 +- ar/hostinger-vps.mdx | 4 +- ar/improve.mdx | 2 +- ar/installation.mdx | 4 +- ar/mcp-integration.mdx | 2 +- ar/memory.mdx | 2 +- ar/migrating-from-ai-sdk.mdx | 2 +- ar/migrating-to-create-agent.mdx | 2 +- ar/models-and-cost.mdx | 2 +- ar/nextjs.mdx | 2 +- ar/oauth.mdx | 2 +- ar/observability.mdx | 2 +- ar/openapi-tools.mdx | 2 +- ar/pause-and-resume.mdx | 2 +- ar/permission-modes.mdx | 2 +- ar/prompting-techniques.mdx | 22 ++-- ar/providers.mdx | 4 +- ar/queue-and-steer.mdx | 2 +- ar/quickstart.mdx | 6 +- ar/react.mdx | 2 +- ar/reasoning.mdx | 2 +- ar/registry.mdx | 4 +- ar/runs.mdx | 2 +- ar/schedules.mdx | 2 +- ar/self-healing-paper.mdx | 4 +- ar/self-healing.mdx | 2 +- ar/sessions.mdx | 2 +- ar/skills.mdx | 2 +- ar/stream-events.mdx | 2 +- ar/streaming.mdx | 2 +- ar/structured-output.mdx | 2 +- ar/sub-agents.mdx | 2 +- ar/svelte.mdx | 2 +- ar/testing.mdx | 2 +- ar/tool-search.mdx | 2 +- ar/tools.mdx | 2 +- ar/translations.json | 156 ++++++++++++------------- ar/triggers.mdx | 2 +- ar/troubleshooting.mdx | 6 +- ar/upgrading.mdx | 2 +- ar/utilities.mdx | 2 +- ar/vue.mdx | 2 +- ar/workspace-tools.mdx | 2 +- auth.mdx | 2 +- build-a-coding-agent.mdx | 2 +- channels.mdx | 2 +- claude-projects.mdx | 2 +- cli.mdx | 2 +- cloudflare-workers.mdx | 2 +- code-mode.mdx | 2 +- coding-agents.mdx | 2 +- compaction.mdx | 2 +- comparison.mdx | 4 +- configuration.mdx | 2 +- connectors.mdx | 2 +- decisions.mdx | 4 +- deployment.mdx | 2 +- docs.json | 4 +- durable-execution.mdx | 2 +- errors.mdx | 2 +- evals.mdx | 2 +- examples.mdx | 34 +++--- examples/coding-agent-workflows.mdx | 2 +- examples/connectors.mdx | 2 +- examples/heal-benchmark.mdx | 2 +- examples/hostinger-deploy.mdx | 4 +- examples/phase-pipeline.mdx | 2 +- examples/waves.mdx | 4 +- executor-api.mdx | 2 +- flows.mdx | 2 +- guardrails.mdx | 4 +- handoffs.mdx | 2 +- hooks.mdx | 2 +- hosted-tools.mdx | 2 +- hostinger-vps.mdx | 4 +- improve.mdx | 2 +- installation.mdx | 4 +- mcp-integration.mdx | 2 +- memory.mdx | 2 +- migrating-from-ai-sdk.mdx | 2 +- migrating-to-create-agent.mdx | 2 +- models-and-cost.mdx | 2 +- nextjs.mdx | 2 +- oauth.mdx | 2 +- observability.mdx | 2 +- openapi-tools.mdx | 2 +- pause-and-resume.mdx | 2 +- permission-modes.mdx | 2 +- prompting-techniques.mdx | 22 ++-- providers.mdx | 4 +- queue-and-steer.mdx | 2 +- quickstart.mdx | 6 +- react.mdx | 2 +- reasoning.mdx | 2 +- registry.mdx | 4 +- runs.mdx | 2 +- schedules.mdx | 2 +- scripts/sync-pr-body.mjs | 4 +- scripts/sync-sdk-docs.mjs | 4 +- self-healing-paper.mdx | 4 +- self-healing.mdx | 2 +- sessions.mdx | 2 +- skills.mdx | 2 +- stream-events.mdx | 2 +- streaming.mdx | 2 +- structured-output.mdx | 2 +- sub-agents.mdx | 2 +- svelte.mdx | 2 +- testing.mdx | 2 +- tool-search.mdx | 2 +- tools.mdx | 2 +- triggers.mdx | 2 +- troubleshooting.mdx | 6 +- upgrading.mdx | 2 +- utilities.mdx | 2 +- vue.mdx | 2 +- workspace-tools.mdx | 2 +- 162 files changed, 336 insertions(+), 336 deletions(-) diff --git a/.github/workflows/sync-sdk-docs.yml b/.github/workflows/sync-sdk-docs.yml index d718d05..c13d5d6 100644 --- a/.github/workflows/sync-sdk-docs.yml +++ b/.github/workflows/sync-sdk-docs.yml @@ -34,7 +34,7 @@ jobs: - uses: actions/checkout@v4 with: - repository: LinuxDevil/agent-sdk + repository: LinuxDevil/lousho ref: main path: sdk diff --git a/README.md b/README.md index 59b3d3a..38412a0 100644 --- a/README.md +++ b/README.md @@ -1,6 +1,6 @@ # Lousho documentation -The documentation site for [`lousho`](https://github.com/LinuxDevil/agent-sdk), +The documentation site for [`lousho`](https://github.com/LinuxDevil/lousho), live at **[lousho.com](https://lousho.com)** in English and [Arabic](https://lousho.com/ar/introduction). Built with [Mintlify](https://mintlify.com). diff --git a/acp.mdx b/acp.mdx index f80bb2a..afd0a97 100644 --- a/acp.mdx +++ b/acp.mdx @@ -2,7 +2,7 @@ title: "Agent Client Protocol (ACP)" sidebarTitle: "Editors (ACP)" --- -{/* Generated from docs/acp.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/acp.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} The [Agent Client Protocol](https://agentclientprotocol.com) lets an editor drive a coding agent that runs as a subprocess: the editor writes JSON-RPC 2.0 diff --git a/agent-directories.mdx b/agent-directories.mdx index de80c7b..542975d 100644 --- a/agent-directories.mdx +++ b/agent-directories.mdx @@ -1,7 +1,7 @@ --- title: "Agent directories" --- -{/* Generated from docs/agent-directories.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/agent-directories.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} An agent can be defined as a directory. `loadAgentDir()` reads it and calls `createAgent()` with the options the files describe, so it returns exactly what diff --git a/agent-forge.mdx b/agent-forge.mdx index caf0eba..5f60db7 100644 --- a/agent-forge.mdx +++ b/agent-forge.mdx @@ -1,7 +1,7 @@ --- title: "Agent Forge" --- -{/* Generated from docs/agent-forge.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/agent-forge.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Agent Forge (`apps/agent-forge`, also called the Studio) is this SDK's companion dashboard: one app with a left navigation to build and run agents, @@ -807,7 +807,7 @@ automatically as long as `apps/agent-forge/dist-server` hasn't been built yet: ```bash -git clone https://github.com/LinuxDevil/agent-sdk.git +git clone https://github.com/LinuxDevil/lousho.git cd agent-sdk npm install npx lousho studio --dev # or just `npx lousho studio` before building diff --git a/agent-spec.mdx b/agent-spec.mdx index f5442a6..cdac3b0 100644 --- a/agent-spec.mdx +++ b/agent-spec.mdx @@ -2,7 +2,7 @@ title: "Agent spec v2" sidebarTitle: "Agent spec v2" --- -{/* Generated from docs/agent-spec.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/agent-spec.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} An agent spec v2 is a YAML or JSON file that describes a whole agent as plain data: model, instructions, tools, MCP servers, skills, memory, sub-agents, handoffs, guardrails, permission rules, hooks, structured output, triggers and an optional flow. Agent Forge, the Agent Author and `lousho dev` all read and write this format. @@ -381,9 +381,9 @@ Agent Forge Templates screen creates an agent from any of them. | Template | What it does | | -------- | ------------ | -| [`templates/support-triage`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/support-triage/spec.yaml) | Sorts each new support ticket by type and urgency and drafts a first reply for a person to send. | -| [`templates/inbox-triage`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/inbox-triage/spec.yaml) | Labels each incoming email, remembers what needs a reply, and emails you a short digest every weekday morning. | -| [`templates/incident-responder`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/incident-responder/spec.yaml) | Turns each firing alert into a short incident summary with likely causes and next steps, and posts it to Slack. | -| [`templates/pr-reviewer`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/pr-reviewer/spec.yaml) | Reviews each new or updated pull request and posts one review comment after you approve it. | -| [`templates/research-assistant`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/research-assistant/spec.yaml) | Answers a research question by reading web pages and citing every source it used. | -| [`templates/docs-qa`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/docs-qa/spec.yaml) | Answers questions about your documentation in chat or Slack and links the pages it used. | +| [`templates/support-triage`](https://github.com/LinuxDevil/lousho/blob/main/templates/support-triage/spec.yaml) | Sorts each new support ticket by type and urgency and drafts a first reply for a person to send. | +| [`templates/inbox-triage`](https://github.com/LinuxDevil/lousho/blob/main/templates/inbox-triage/spec.yaml) | Labels each incoming email, remembers what needs a reply, and emails you a short digest every weekday morning. | +| [`templates/incident-responder`](https://github.com/LinuxDevil/lousho/blob/main/templates/incident-responder/spec.yaml) | Turns each firing alert into a short incident summary with likely causes and next steps, and posts it to Slack. | +| [`templates/pr-reviewer`](https://github.com/LinuxDevil/lousho/blob/main/templates/pr-reviewer/spec.yaml) | Reviews each new or updated pull request and posts one review comment after you approve it. | +| [`templates/research-assistant`](https://github.com/LinuxDevil/lousho/blob/main/templates/research-assistant/spec.yaml) | Answers a research question by reading web pages and citing every source it used. | +| [`templates/docs-qa`](https://github.com/LinuxDevil/lousho/blob/main/templates/docs-qa/spec.yaml) | Answers questions about your documentation in chat or Slack and links the pages it used. | diff --git a/ai-sdk-ui.mdx b/ai-sdk-ui.mdx index 7305149..20dee65 100644 --- a/ai-sdk-ui.mdx +++ b/ai-sdk-ui.mdx @@ -2,7 +2,7 @@ title: "AI SDK UI (useChat)" sidebarTitle: "AI SDK UI (useChat)" --- -{/* Generated from docs/ai-sdk-ui.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/ai-sdk-ui.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} If your front end already uses the Vercel AI SDK's UI hooks (`useChat` from `@ai-sdk/react`, and the Vue, Svelte and Angular equivalents), you can keep diff --git a/api-overview.mdx b/api-overview.mdx index 1d1496a..3052a56 100644 --- a/api-overview.mdx +++ b/api-overview.mdx @@ -3,7 +3,7 @@ title: "API Overview" sidebarTitle: "API overview" description: "The package root (lousho) exports everything below. For the complete, generated API reference (every export, signature and doc comment), build the TypeDoc site:" --- -{/* Generated from docs/api-overview.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/api-overview.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ```bash npm run docs:build # writes docs/api/index.html diff --git a/approvals.mdx b/approvals.mdx index f2f45de..0bc6de7 100644 --- a/approvals.mdx +++ b/approvals.mdx @@ -1,7 +1,7 @@ --- title: "Approvals" --- -{/* Generated from docs/approvals.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/approvals.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Flag a tool `needsApproval` and the run pauses before calling it, until a human (or your code) approves or rejects the call. The pause is saved in an diff --git a/ar/acp.mdx b/ar/acp.mdx index 69e0d36..8c9bc54 100644 --- a/ar/acp.mdx +++ b/ar/acp.mdx @@ -2,7 +2,7 @@ title: "Agent Client Protocol (ACP)" sidebarTitle: "المحررات (ACP)" --- -{/* Generated from docs/acp.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/acp.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} [Agent Client Protocol](https://agentclientprotocol.com) بروتوكول يتيح للمحرر أن يقود agent للبرمجة يعمل كعملية فرعية: المحرر يكتب طلبات JSON-RPC 2.0 diff --git a/ar/agent-directories.mdx b/ar/agent-directories.mdx index 05e4818..86cfb4f 100644 --- a/ar/agent-directories.mdx +++ b/ar/agent-directories.mdx @@ -1,7 +1,7 @@ --- title: "مجلدات agent" --- -{/* Generated from docs/agent-directories.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/agent-directories.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} يمكنك تعريف الـ agent كمجلد. `loadAgentDir()` تقرأه وتستدعي `createAgent()` بالخيارات التي تصفها الملفات، فتُرجع بالضبط ما تُرجعه `createAgent()`. يبقى الـ typed code API هو المرجع الأساسي: المجلد طريقة أخرى لكتابة الخيارات نفسها، ويمكنك الانتقال إلى الكود في أي وقت دون إعادة كتابة. diff --git a/ar/agent-forge.mdx b/ar/agent-forge.mdx index bdf0e44..6697ab4 100644 --- a/ar/agent-forge.mdx +++ b/ar/agent-forge.mdx @@ -1,7 +1,7 @@ --- title: "Agent Forge" --- -{/* Generated from docs/agent-forge.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/agent-forge.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Agent Forge (`apps/agent-forge`، واسمه الآخر الـ Studio) هو الـ dashboard المرافق لهذه الـ SDK: تطبيق واحد فيه navigation على اليسار تبني منه الـ @@ -793,7 +793,7 @@ Forge ذاته، لا تستخدمه فقط — فإن `lousho studio` يعود `apps/agent-forge/dist-server` لم يُبنَ بعد: ```bash -git clone https://github.com/LinuxDevil/agent-sdk.git +git clone https://github.com/LinuxDevil/lousho.git cd agent-sdk npm install npx lousho studio --dev # or just `npx lousho studio` before building diff --git a/ar/agent-spec.mdx b/ar/agent-spec.mdx index 044b26b..7f29200 100644 --- a/ar/agent-spec.mdx +++ b/ar/agent-spec.mdx @@ -2,7 +2,7 @@ title: "Agent spec v2" sidebarTitle: "Agent spec v2" --- -{/* Generated from docs/agent-spec.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/agent-spec.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ملف مواصفات الـ agent بالإصدار 2 (agent spec v2) هو ملف YAML أو JSON يصف agent كاملًا كبيانات عادية: الـ model، والتعليمات، والـ tools، وخوادم MCP، والـ skills، والذاكرة، والـ sub-agents، والـ handoffs، والـ guardrails، وقواعد الصلاحيات، والـ hooks، والمخرجات المنظَّمة، والـ triggers، و flow اختياري. Agent Forge والـ Agent Author و`lousho dev` كلها تقرأ هذه الصيغة وتكتبها. @@ -375,9 +375,9 @@ Templates في Agent Forge تنشئ agent من أي منها. | القالب | ما يفعله | | -------- | ------------ | -| [`templates/support-triage`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/support-triage/spec.yaml) | يفرز كل تذكرة دعم جديدة حسب النوع والأولوية ويكتب مسودة رد أول يرسلها شخص. | -| [`templates/inbox-triage`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/inbox-triage/spec.yaml) | يصنّف كل بريد وارد، ويتذكّر ما يحتاج إلى رد، ويرسل لك ملخصًا قصيرًا كل صباح يوم عمل. | -| [`templates/incident-responder`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/incident-responder/spec.yaml) | يحوّل كل تنبيه يُطلَق إلى ملخص حادثة قصير مع الأسباب المحتملة والخطوات التالية، وينشره في Slack. | -| [`templates/pr-reviewer`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/pr-reviewer/spec.yaml) | يراجع كل pull request جديد أو محدَّث وينشر تعليق مراجعة واحدًا بعد موافقتك عليه. | -| [`templates/research-assistant`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/research-assistant/spec.yaml) | يجيب عن سؤال بحثي بقراءة صفحات الويب ويستشهد بكل مصدر استخدمه. | -| [`templates/docs-qa`](https://github.com/LinuxDevil/agent-sdk/blob/main/templates/docs-qa/spec.yaml) | يجيب عن أسئلة حول توثيقك في المحادثة أو Slack ويضع روابط الصفحات التي استخدمها. | +| [`templates/support-triage`](https://github.com/LinuxDevil/lousho/blob/main/templates/support-triage/spec.yaml) | يفرز كل تذكرة دعم جديدة حسب النوع والأولوية ويكتب مسودة رد أول يرسلها شخص. | +| [`templates/inbox-triage`](https://github.com/LinuxDevil/lousho/blob/main/templates/inbox-triage/spec.yaml) | يصنّف كل بريد وارد، ويتذكّر ما يحتاج إلى رد، ويرسل لك ملخصًا قصيرًا كل صباح يوم عمل. | +| [`templates/incident-responder`](https://github.com/LinuxDevil/lousho/blob/main/templates/incident-responder/spec.yaml) | يحوّل كل تنبيه يُطلَق إلى ملخص حادثة قصير مع الأسباب المحتملة والخطوات التالية، وينشره في Slack. | +| [`templates/pr-reviewer`](https://github.com/LinuxDevil/lousho/blob/main/templates/pr-reviewer/spec.yaml) | يراجع كل pull request جديد أو محدَّث وينشر تعليق مراجعة واحدًا بعد موافقتك عليه. | +| [`templates/research-assistant`](https://github.com/LinuxDevil/lousho/blob/main/templates/research-assistant/spec.yaml) | يجيب عن سؤال بحثي بقراءة صفحات الويب ويستشهد بكل مصدر استخدمه. | +| [`templates/docs-qa`](https://github.com/LinuxDevil/lousho/blob/main/templates/docs-qa/spec.yaml) | يجيب عن أسئلة حول توثيقك في المحادثة أو Slack ويضع روابط الصفحات التي استخدمها. | diff --git a/ar/ai-sdk-ui.mdx b/ar/ai-sdk-ui.mdx index b77c757..98b5af1 100644 --- a/ar/ai-sdk-ui.mdx +++ b/ar/ai-sdk-ui.mdx @@ -2,7 +2,7 @@ title: "AI SDK UI (useChat)" sidebarTitle: "AI SDK UI (useChat)" --- -{/* Generated from docs/ai-sdk-ui.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/ai-sdk-ui.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} إذا كانت الواجهة الأمامية (front end) لديك تستخدم أصلًا الـ UI hooks في Vercel AI SDK (`useChat` من `@ai-sdk/react` ومثيلاتها في Vue وSvelte وAngular)، فيمكنك إبقاؤها على حالها وتشغيل Lousho agent خلف الـ route. ثلاث دوال مُصدَّرة من جذر الحزمة تتولى التحويل. لا تستورد شيئًا من `ai`، ولا تستخدم أي `node:*` module، وتعمل على Node وWorkers وأي runtime مبني على Fetch. diff --git a/ar/api-overview.mdx b/ar/api-overview.mdx index 72c047d..28a5aa8 100644 --- a/ar/api-overview.mdx +++ b/ar/api-overview.mdx @@ -3,7 +3,7 @@ title: "نظرة عامة على الـ API" sidebarTitle: "نظرة عامة على الـ API" description: "جذر الحزمة (lousho) يصدّر كل شيء أدناه. للمرجع الكامل المولّد للـ API (كل export وكل signature وكل تعليق توثيقي)، ابنِ موقع TypeDoc:" --- -{/* Generated from docs/api-overview.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/api-overview.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ```bash npm run docs:build # writes docs/api/index.html diff --git a/ar/approvals.mdx b/ar/approvals.mdx index 703b298..1c449e3 100644 --- a/ar/approvals.mdx +++ b/ar/approvals.mdx @@ -1,7 +1,7 @@ --- title: "Approvals" --- -{/* Generated from docs/approvals.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/approvals.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} علّم الـ tool بـ `needsApproval` فيتوقف الـ run قبل تنفيذ الاستدعاء، حتى يوافق إنسان (أو الكود الخاص بك) على الـ call أو يرفضه. التوقف محفوظ diff --git a/ar/auth.mdx b/ar/auth.mdx index 538a2ec..1b50b79 100644 --- a/ar/auth.mdx +++ b/ar/auth.mdx @@ -1,7 +1,7 @@ --- title: "Route auth والـ principals" --- -{/* Generated from docs/auth.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/auth.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} أي agent يعمل عبر HTTP يحتاج إلى معرفة أمرين عن كل request: هل يُسمح له بالمرور أصلًا، ومن المتصل. تجيب `lousho/auth` عن السؤالين. أنت تعطي الـ route قائمة مرتبة من مدخلات auth: `jwt()` أو `oidc()` أو `basic()` أو `apiToken()` أو `anonymous()`، أو function تكتبها بنفسك. أول entry يقبل الـ request يُرجع **principal** (هوية المتصل بعد أن تحققت منها auth)، وهذا الـ principal يدخل الـ run، وهناك تستطيع دوال `model` و`instructions` و`tools` ونطاقات الـ memory قراءته. diff --git a/ar/build-a-coding-agent.mdx b/ar/build-a-coding-agent.mdx index b013a35..6ac4dcc 100644 --- a/ar/build-a-coding-agent.mdx +++ b/ar/build-a-coding-agent.mdx @@ -2,7 +2,7 @@ title: "ابنِ coding agent" sidebarTitle: "ابنِ coding agent" --- -{/* Generated from docs/build-a-coding-agent.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/build-a-coding-agent.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} هذه الصفحة تبني coding agent يعمل في الـ terminal على طريقة Claude Code: يقرأ ملفات مشروعك ويعدّلها، ويشغّل أوامر، ويسألك قبل أن يكتب ملفًا أو يشغّل شيئًا خطرًا، ويُبقي المحادثة مستمرة لأسئلة المتابعة. تحتاج إلى Node 22، ومفتاح API في متغيّر البيئة الذي يتوقعه الـ provider (`OPENAI_API_KEY` للـ model المستخدم هنا)، ومجلد مشروع تشغّله فيه. الملف الجاهز في آخر [الخطوة 5](#5-حافظ-على-استمرار-المحادثة)، وتشغّله هكذا `npx tsx coding.ts "Fix the failing test"`. diff --git a/ar/channels.mdx b/ar/channels.mdx index 73b9734..ab11b92 100644 --- a/ar/channels.mdx +++ b/ar/channels.mdx @@ -1,7 +1,7 @@ --- title: "Channels" --- -{/* Generated from docs/channels.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/channels.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} يربط الـ **channel** الـ agent بطرف واحد: JSON API، أو webhook، أو تطبيق محادثة. وهو الذي يحدد طريقة مصادقة الطلب الوارد، والمحادثة التي ينتمي إليها، وكيف diff --git a/ar/claude-projects.mdx b/ar/claude-projects.mdx index 76da382..de7d0c2 100644 --- a/ar/claude-projects.mdx +++ b/ar/claude-projects.mdx @@ -1,7 +1,7 @@ --- title: "مشاريع Claude Code" --- -{/* Generated from docs/claude-projects.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/claude-projects.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} المستودع الذي يحمل أصلًا ملفات Claude Code المعتادة — `CLAUDE.md` و`.claude/skills/` و`.claude/agents/` و`.claude/rules/` diff --git a/ar/cli.mdx b/ar/cli.mdx index 6238538..76565bb 100644 --- a/ar/cli.mdx +++ b/ar/cli.mdx @@ -1,7 +1,7 @@ --- title: "CLI" --- -{/* Generated from docs/cli.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/cli.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} تثبيت الحزمة يثبّت معها أمر `lousho`. ينشئ مشاريع جديدة، ويفحص الإعداد الخاص بك، ويشغّل agent محليًا، ويقدّمه عبر MCP، ويشغّل evals، ويعرض traces محفوظة، ويبني artifact جاهزًا للنشر، ويفتح لوحة Agent Forge. شغّله بـ `npx lousho ` داخل مشروع مثبّت فيه الـ SDK. diff --git a/ar/cloudflare-workers.mdx b/ar/cloudflare-workers.mdx index 78d3623..b63e0c7 100644 --- a/ar/cloudflare-workers.mdx +++ b/ar/cloudflare-workers.mdx @@ -2,7 +2,7 @@ title: "Cloudflare Workers" --- -{/* Generated from docs/cloudflare-workers.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/cloudflare-workers.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الأمر `npx lousho build --target=cloudflare-worker --agent=agent.yaml` يولّد module Worker باسم `worker.ts`، و`wrangler.toml`، وbundle هو `dist/worker.js` يفحصه الـ build بحثًا عن استيرادات `node:`. يحتاج `tsup` و`wrangler` مثبتين، ويأخذ [ملف agent spec](/ar/configuration) أو [agent directory](/ar/agent-directories) (`npx lousho build ./my-agent --target=cloudflare-worker`). الـ Worker يقدّم نفس [HTTP API](/ar/deployment#http-api) التي تقدّمها الأهداف الثانية. هذا الهدف هو الأضيق حدودًا بين الثلاثة، فنبدأ بحدوده. diff --git a/ar/code-mode.mdx b/ar/code-mode.mdx index 3954091..2f751bf 100644 --- a/ar/code-mode.mdx +++ b/ar/code-mode.mdx @@ -1,7 +1,7 @@ --- title: "Code mode" --- -{/* Generated from docs/code-mode.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/code-mode.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} مع code mode يحصل الموديل على tool إضافية: `run_code`. بدلًا من أن يستدعي الـ tools واحدة تلو الأخرى، مع round trip للموديل بعد كل call، يكتب برنامج JavaScript قصيرًا diff --git a/ar/coding-agents.mdx b/ar/coding-agents.mdx index b32ee6a..469f8eb 100644 --- a/ar/coding-agents.mdx +++ b/ar/coding-agents.mdx @@ -10,7 +10,7 @@ description: "أعطِ Claude Code أو Cursor أو أي coding agent آخر ت | الملف | أين يوجد | ما هو | | ---- | ----- | ---------- | | `llms.txt` | في حزمة npm: `node_modules/lousho/llms.txt` | فهرس [llmstxt.org](https://llmstxt.org): سطر واحد لكل صفحة توثيق، مع ملخص ورابط | -| `llms-full.txt` | في [مستودع الـ SDK](https://github.com/LinuxDevil/agent-sdk/blob/main/llms-full.txt)، وليس في حزمة npm (حجمه حوالي 1 MB) | الـ README وكل صفحات التوثيق في ملف واحد، بروابط مطلقة | +| `llms-full.txt` | في [مستودع الـ SDK](https://github.com/LinuxDevil/lousho/blob/main/llms-full.txt)، وليس في حزمة npm (حجمه حوالي 1 MB) | الـ README وكل صفحات التوثيق في ملف واحد، بروابط مطلقة | الملفان مولّدان من نفس توثيق هذا الموقع، والـ CI الخاص بالـ SDK يتحقق منهما. ملف `llms.txt` الموجود في `node_modules` يطابق النسخة المثبتة عندك؛ أما `llms-full.txt` في المستودع فيتبع فرع `main` في الـ SDK، وقد يكون أحدث من الإصدار الذي ثبّتّه. أخبر الـ coding agent بمكانهما، مثلًا في `AGENTS.md` أو `CLAUDE.md` في مشروعك: diff --git a/ar/compaction.mdx b/ar/compaction.mdx index df7cd55..0fb9a82 100644 --- a/ar/compaction.mdx +++ b/ar/compaction.mdx @@ -1,7 +1,7 @@ --- title: "Context compaction" --- -{/* Generated from docs/compaction.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/compaction.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} run طويل لـagent يملأ الـcontext window غالبًا بنتائج الـtools: محتويات ملفات، نتائج بحث، ردود API قرأها الـmodel مرة ولم يعد يحتاجها كاملة. الـcompaction (ضغط سياق المحادثة) يُبقي run كهذا تحت حد الـmodel على مرحلتين: أولاً يستبدل بنتائج الـtools القديمة علامة قصيرة، وإذا لم يكفِ ذلك يستبدل بأقدم الدورات ملخصًا يكتبه model (أرخص). diff --git a/ar/comparison.mdx b/ar/comparison.mdx index a2d5a83..95ba779 100644 --- a/ar/comparison.mdx +++ b/ar/comparison.mdx @@ -2,10 +2,10 @@ title: "مقارنة Lousho بغيرها" description: "تقارن هذه الصفحة Lousho بأربع مكتبات وكلاء أخرى مكتوبة بـ TypeScript: Vercel AI SDK (ToolLoopAgent)، وOpenAI Agents SDK للغة JavaScript، وMastra، وLangGraph.js. كل خانة تخص مشروعًا آخر مأخوذة من توثيقه الرسمي." --- -{/* Generated from docs/comparison.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/comparison.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} آخر تحقق: 9 نوفمبر 2026. هذه المشاريع تتغير بسرعة، لذا قد تكون بعض الخانات قديمة. -نرحب بالتصحيحات: [افتح issue](https://github.com/LinuxDevil/agent-sdk/issues). +نرحب بالتصحيحات: [افتح issue](https://github.com/LinuxDevil/lousho/issues). ## كيف تقرأ الجدول diff --git a/ar/configuration.mdx b/ar/configuration.mdx index 777e8a9..ea40a09 100644 --- a/ar/configuration.mdx +++ b/ar/configuration.mdx @@ -1,7 +1,7 @@ --- title: "الإعدادات" --- -{/* Generated from docs/configuration.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/configuration.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## ملفات تعريف الـ agent (`AgentSpec`) diff --git a/ar/connectors.mdx b/ar/connectors.mdx index 78c15be..4962291 100644 --- a/ar/connectors.mdx +++ b/ar/connectors.mdx @@ -1,7 +1,7 @@ --- title: "الموصِّلات (Connectors)" --- -{/* Generated from docs/connectors.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/connectors.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ connectors (الموصِّلات) هي طريقة الـ agent للوصول إلى خدمات خارجية، مثل Google Docs و Slack و GitHub و Notion أو قاعدة بيانات، بنفس طريقة connectors في Claude. في Lousho الـ diff --git a/ar/decisions.mdx b/ar/decisions.mdx index a5e1caf..65dbceb 100644 --- a/ar/decisions.mdx +++ b/ar/decisions.mdx @@ -1,7 +1,7 @@ --- title: "Decisions" --- -{/* Generated from docs/decisions.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/decisions.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `decide()` يستدعي OpenAI Decisions API (`POST /v1/decisions`): endpoint مصمَّم لهذا الغرض، يقيّم input مشتركًا مقابل أسئلة typed ويرجّع إجابات معايَرة (calibrated) أسرع بحوالي 10 مرات من دورة model كاملة. هو الشكل السريع والحتمي لنمط القرار الـ typed، أي التصنيف والتوجيه والبوابات والتقييم بالدرجات، دون أن تشغّل run كاملًا لـ agent. @@ -114,4 +114,4 @@ const result = await decide({ - **[المخرجات المنظَّمة](/ar/structured-output)** (`createAgent({ output })`): حين تحتاج JSON schema خاصًا بك: حقول مستخرجة، أو شرح مكتوب، أو object على شكل الـ domain عندك. يعمل مع أي provider، بما فيها endpoints المخرجات المنظَّمة الصارمة (راجع [حدود الـ schema](/ar/structured-output#حدود-الـ-schema)). - **[الوكلاء الفرعيون](/ar/sub-agents)** أو [مسارات العمل](/ar/flows): حين يحتاج القرار tools أو استدلالًا متعدد الخطوات قبل أن يجيب. -[مثال coding-agent-workflows](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/coding-agent-workflows) يعرض نفس نمط التوجيه/الثقة/الحد الأدنى مبنيًا على schemas الخاصة بـ `output`؛ استبدل خطوة الفرز فيه بـ `decide()` إذا كنت على OpenAI وتريد الـ endpoint المخصص. وصفحة [تقنيات كتابة الموجّهات](/ar/prompting-techniques#قرارات-typed-بأسلوب-system-one) تضع الشكلين في سياق كتالوج القرارات الـ typed الأوسع. +[مثال coding-agent-workflows](https://github.com/LinuxDevil/lousho/tree/main/examples/coding-agent-workflows) يعرض نفس نمط التوجيه/الثقة/الحد الأدنى مبنيًا على schemas الخاصة بـ `output`؛ استبدل خطوة الفرز فيه بـ `decide()` إذا كنت على OpenAI وتريد الـ endpoint المخصص. وصفحة [تقنيات كتابة الموجّهات](/ar/prompting-techniques#قرارات-typed-بأسلوب-system-one) تضع الشكلين في سياق كتالوج القرارات الـ typed الأوسع. diff --git a/ar/deployment.mdx b/ar/deployment.mdx index a9dc3f2..8d2d704 100644 --- a/ar/deployment.mdx +++ b/ar/deployment.mdx @@ -1,7 +1,7 @@ --- title: "النشر" --- -{/* Generated from docs/deployment.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/deployment.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} يأخذ `lousho build` ملف spec للـagent (راجع [الإعداد](/ar/configuration)) ويُخرج منه artifact جاهزًا للنشر على منصة واحدة: diff --git a/ar/durable-execution.mdx b/ar/durable-execution.mdx index 9c9b5e5..addedfe 100644 --- a/ar/durable-execution.mdx +++ b/ar/durable-execution.mdx @@ -1,7 +1,7 @@ --- title: "Durable execution" --- -{/* Generated from docs/durable-execution.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/durable-execution.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} مرّر `store` لـ`createAgent()` ومرّر `sessionId` لـ`send()`، فيستمر الـ run رغم crash، أو restart، أو abort، أو approval (موافقة) من إنسان تستغرق أيامًا. ولا شيء هنا يعتمد على host معيّن: أي `CheckpointStore` — `memoryStore()`، أو `SqliteStore`، أو store الـ Cloudflare KV الذي يوفّره build `cloudflare-worker`، أو تنفيذك الخاص بثلاث دوال — وأي `ApprovalStore` تعمل بالطريقة نفسها. diff --git a/ar/errors.mdx b/ar/errors.mdx index 61b2b15..96a9e81 100644 --- a/ar/errors.mdx +++ b/ar/errors.mdx @@ -3,7 +3,7 @@ title: "الأخطاء" description: "كل خطأ يرميه الـ SDK عن قصد هو SDKError بكود LOUSHO_* ثابت وتلميح ورابط لقسمه هنا؛ تسرد هذه الصفحة كل كود مع سببه وحلّه." --- -{/* Generated from docs/errors.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/errors.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} تقريبًا كل خطأ يرميه الـ SDK عن قصد هو `SDKError` (أو subclass منه مثل `ConfigurationError` أو `ValidationError` أو `MissingPeerDependencyError`) وفيه: diff --git a/ar/evals.mdx b/ar/evals.mdx index e578454..be90dc4 100644 --- a/ar/evals.mdx +++ b/ar/evals.mdx @@ -1,7 +1,7 @@ --- title: "Evals" --- -{/* Generated from docs/evals.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/evals.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ evals هي regression tests لـ *سلوك* الـ agent: أي tools استدعى، وبأي ترتيب، وبأي arguments، وكم خطوة أخذ، وماذا قال. diff --git a/ar/examples.mdx b/ar/examples.mdx index d7ae2b0..de417b8 100644 --- a/ar/examples.mdx +++ b/ar/examples.mdx @@ -3,17 +3,17 @@ title: "أمثلة" description: "أمثلة جاهزة للتشغيل من مستودع الـ SDK. معظمها يعمل على mock provider مدمج، فلا تحتاج مفتاح API." --- -الأمثلة موجودة في مجلد [`examples/`](https://github.com/LinuxDevil/agent-sdk/tree/main/examples) داخل مستودع الـ SDK. اعمل clone للمستودع، ثبّت الاعتماديات، وشغّل أي مثال بالأمر المكتوب بجانبه: +الأمثلة موجودة في مجلد [`examples/`](https://github.com/LinuxDevil/lousho/tree/main/examples) داخل مستودع الـ SDK. اعمل clone للمستودع، ثبّت الاعتماديات، وشغّل أي مثال بالأمر المكتوب بجانبه: ```bash -git clone https://github.com/LinuxDevil/agent-sdk.git +git clone https://github.com/LinuxDevil/lousho.git cd agent-sdk npm install ``` ## المثال الأبرز - + تنبيه مراقبة يصل إلى monitor agent، فيفوّض الإصلاح خلف بوابة approval (موافقة) بشرية. زر "Fix it" في Slack يستأنف الـ run المتوقف، وfixer agent يكتب patch، والـ guardrails تفحص الـ patch قبل فتح pull request على GitHub. يعمل على mocks افتراضيًا: `npm run pipeline:demo`. @@ -21,25 +21,25 @@ npm install | المثال | ما يعرضه | الأمر | | ------- | ------------- | ------ | -| [coding-harness](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/coding-harness) | coding agent بمواصفات production: أدوات workspace مع checkpoints، قواعد صلاحيات، loop guard، وسقف تكلفة، ويُشحن على شكل registry kits باسم `coding-kit` و`coding-pi` | `npx tsx examples/coding-harness/index.ts` | -| [deep-research](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/deep-research) | نمط الـ multi-agent للبحث: coordinator يوزّع استدعاءات `task` متوازية على researcher sub-agents، ثم يجمّع تقريرًا موثّقًا بالمصادر | `npm run example:deep-research` | -| [support-desk](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/support-desk) | agent للفرز يحوّل المحادثة بـ `handoff()` إلى متخصصين في الفوترة والتقنية، مع أداة refund محمية بـ approval | `npm run example:support-desk` | -| [evaluator-loop](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/evaluator-loop) | نمط الكاتب/الناقد: `llmCritique()` يقيّم المسودة، والكاتب يراجعها حسب ملاحظاته حتى تنجح | `npm run example:evaluator-loop` | -| [data-analyst](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/data-analyst) | تحويل نص إلى SQL فوق `node:sqlite` مع فرض قراءة-فقط على أكثر من طبقة — يستحيل أن تعمل `DELETE`/`DROP` | `npm run example:data-analyst` | -| [plan-mode](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/plan-mode) | coding agent يقرأ ويقترح في وضع `plan`، ثم يطبّق التعديلات بعد `session.setPermissionMode('acceptEdits')` | `npm run example:plan-mode` | +| [coding-harness](https://github.com/LinuxDevil/lousho/tree/main/examples/coding-harness) | coding agent بمواصفات production: أدوات workspace مع checkpoints، قواعد صلاحيات، loop guard، وسقف تكلفة، ويُشحن على شكل registry kits باسم `coding-kit` و`coding-pi` | `npx tsx examples/coding-harness/index.ts` | +| [deep-research](https://github.com/LinuxDevil/lousho/tree/main/examples/deep-research) | نمط الـ multi-agent للبحث: coordinator يوزّع استدعاءات `task` متوازية على researcher sub-agents، ثم يجمّع تقريرًا موثّقًا بالمصادر | `npm run example:deep-research` | +| [support-desk](https://github.com/LinuxDevil/lousho/tree/main/examples/support-desk) | agent للفرز يحوّل المحادثة بـ `handoff()` إلى متخصصين في الفوترة والتقنية، مع أداة refund محمية بـ approval | `npm run example:support-desk` | +| [evaluator-loop](https://github.com/LinuxDevil/lousho/tree/main/examples/evaluator-loop) | نمط الكاتب/الناقد: `llmCritique()` يقيّم المسودة، والكاتب يراجعها حسب ملاحظاته حتى تنجح | `npm run example:evaluator-loop` | +| [data-analyst](https://github.com/LinuxDevil/lousho/tree/main/examples/data-analyst) | تحويل نص إلى SQL فوق `node:sqlite` مع فرض قراءة-فقط على أكثر من طبقة — يستحيل أن تعمل `DELETE`/`DROP` | `npm run example:data-analyst` | +| [plan-mode](https://github.com/LinuxDevil/lousho/tree/main/examples/plan-mode) | coding agent يقرأ ويقترح في وضع `plan`، ثم يطبّق التعديلات بعد `session.setPermissionMode('acceptEdits')` | `npm run example:plan-mode` | | [waves](/ar/examples/waves) | الموجات (waves): عمّال متوازيون بعدد محدود، وتجميع حتمي، وبوابة تحقّق مُنمَّطة، وموجات توسعة تنطلق من الثغرات التي تسمّيها تلك البوابة | `npm run example:waves` | | [phase-pipeline](/ar/examples/phase-pipeline) | هندسة المراحل فوق الـ flows: نقاش → spec → خطة → تنفيذ → بوابة إثبات، مع سُلّم إعادة المحاولة على شكل سلسلة `oneOf` | `npm run example:phase-pipeline` | | [coding-agent-workflows](/ar/examples/coding-agent-workflows) | coding agent يُوجَّه بقرار مُنمَّط (`{ route, confidence }` عبر `output`)، مع حدّ أدنى للثقة يصعّد الطلب إلى المسار الحذر | `npm run example:coding-agent-workflows` | | [connectors](/ar/examples/connectors) | MCP connectors (بشكل Claude connectors نفسه) تعمل بالكامل دون اتصال على خادم stdio وهمي مبني بـ `serveMcp` | `npm run example:connectors` | | [hostinger-deploy](/ar/examples/hostinger-deploy) | نشر agent على Hostinger VPS حقيقي: تحقّق عبر الـ API، و`docker save` عبر SSH، واختبارات smoke للـ health والـ approval | `node examples/hostinger-deploy/deploy.mjs --vm ` | -| [agent-dir](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/agent-dir) | agent معرّف كمجلد (`instructions.md` و`tools/` و`skills/` و`agent.json`) ويُحمَّل بـ `loadAgentDir()` | `npm run example:agent-dir` | -| [support-bot](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/support-bot) | أبسط agent لدعم العملاء | `npm run example:support-bot` | -| [research-assistant](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/research-assistant) | agent للبحث يستخدم أداة `http` المدمجة | `npm run example:research-assistant` | -| [doc-qa](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/doc-qa) | الرد على الأسئلة في حدود مستند واحد | `npm run example:doc-qa` | -| [workflow-router](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/workflow-router) | يصنّف الطلب ضمن مجموعة ثابتة من الفئات | `npm run example:workflow-router` | -| [slack-notifier](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/slack-notifier) | يحوّل وصف حدث خام إلى إشعار Slack قصير | `npm run example:slack-notifier` | -| [tracing](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/tracing) | run موصول بمصدّرات trace إلى الـ console وإلى OpenTelemetry | `npm run example:tracing:console` | -| [openrouter](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/openrouter) | `OpenRouterProvider`: توليد، streaming، tool calling، وعرض قائمة الـ models. يحتاج `OPENROUTER_API_KEY` | `npm run example:openrouter` | +| [agent-dir](https://github.com/LinuxDevil/lousho/tree/main/examples/agent-dir) | agent معرّف كمجلد (`instructions.md` و`tools/` و`skills/` و`agent.json`) ويُحمَّل بـ `loadAgentDir()` | `npm run example:agent-dir` | +| [support-bot](https://github.com/LinuxDevil/lousho/tree/main/examples/support-bot) | أبسط agent لدعم العملاء | `npm run example:support-bot` | +| [research-assistant](https://github.com/LinuxDevil/lousho/tree/main/examples/research-assistant) | agent للبحث يستخدم أداة `http` المدمجة | `npm run example:research-assistant` | +| [doc-qa](https://github.com/LinuxDevil/lousho/tree/main/examples/doc-qa) | الرد على الأسئلة في حدود مستند واحد | `npm run example:doc-qa` | +| [workflow-router](https://github.com/LinuxDevil/lousho/tree/main/examples/workflow-router) | يصنّف الطلب ضمن مجموعة ثابتة من الفئات | `npm run example:workflow-router` | +| [slack-notifier](https://github.com/LinuxDevil/lousho/tree/main/examples/slack-notifier) | يحوّل وصف حدث خام إلى إشعار Slack قصير | `npm run example:slack-notifier` | +| [tracing](https://github.com/LinuxDevil/lousho/tree/main/examples/tracing) | run موصول بمصدّرات trace إلى الـ console وإلى OpenTelemetry | `npm run example:tracing:console` | +| [openrouter](https://github.com/LinuxDevil/lousho/tree/main/examples/openrouter) | `OpenRouterProvider`: توليد، streaming، tool calling، وعرض قائمة الـ models. يحتاج `OPENROUTER_API_KEY` | `npm run example:openrouter` | ## أدلة ذات صلة diff --git a/ar/examples/coding-agent-workflows.mdx b/ar/examples/coding-agent-workflows.mdx index 7441c39..e1b4a5c 100644 --- a/ar/examples/coding-agent-workflows.mdx +++ b/ar/examples/coding-agent-workflows.mdx @@ -3,7 +3,7 @@ title: "مسارات عمل الـ coding agent (coding-agent-workflows)" sidebarTitle: "مسارات عمل الـ coding agent" description: "coding agent تقوده مسارات عمل حتمية يكون توجيهها قرارًا مُنمَّطًا (نمط \"System One\")." --- -{/* Generated from examples/coding-agent-workflows/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/coding-agent-workflows/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} يجيب القرار المُنمَّط (typed decision) عن سؤال احتمالي منظَّم، مثل التصنيف أو التوجيه أو التقييم أو التحقق، بدلًا من توليد نص. والصيغة التي يأخذها ذلك في Lousho هي `createAgent({ output: DECISION })`: يُرجع النموذج `{ route, confidence, reason }` بعد التحقق منه بمخطط zod (ويمكن قراءته عبر `result.object`)، ويتفرّع مسار العمل بحسب الحقول المُنمَّطة. diff --git a/ar/examples/connectors.mdx b/ar/examples/connectors.mdx index 2cac191..1f6b4d9 100644 --- a/ar/examples/connectors.mdx +++ b/ar/examples/connectors.mdx @@ -3,7 +3,7 @@ title: "الموصِّلات (connectors)" sidebarTitle: "الموصِّلات" description: "اربط الـ agent بخدمات خارجية (Google Docs وSlack وGitHub وNotion وغيرها) بالطريقة التي تعمل بها موصِّلات Claude: عبر MCP." --- -{/* Generated from examples/connectors/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/connectors/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ```ts const agent = createAgent({ diff --git a/ar/examples/heal-benchmark.mdx b/ar/examples/heal-benchmark.mdx index e81ee52..09dcce5 100644 --- a/ar/examples/heal-benchmark.mdx +++ b/ar/examples/heal-benchmark.mdx @@ -3,7 +3,7 @@ title: "مقياس الإصلاح الذاتي (heal-benchmark)" sidebarTitle: "مقياس الإصلاح الذاتي" description: "طريقة قابلة للتكرار لمعرفة ما إذا كان تغيير في الـ harness الخاص بالـ agent قد ساعد. تشغّل أربعة agents على 28 مهمة وتُبلغ عن معدل النجاح والضجيج والـ tokens واستدعاءات الأدوات لكل نسخة من الـ harness." --- -{/* Generated from examples/heal-benchmark/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/heal-benchmark/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} | الـ agent | ما يختبره | | --- | --- | diff --git a/ar/examples/hostinger-deploy.mdx b/ar/examples/hostinger-deploy.mdx index f523177..b173d9e 100644 --- a/ar/examples/hostinger-deploy.mdx +++ b/ar/examples/hostinger-deploy.mdx @@ -2,7 +2,7 @@ title: "النشر على Hostinger (hostinger-deploy)" sidebarTitle: "النشر على Hostinger" --- -{/* Generated from examples/hostinger-deploy/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/hostinger-deploy/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} انشر agent من Lousho على خادم VPS حقيقي: المجلد `agent/` هو مجلد agent لمساعد عمليات صغير (`vps-ops`)، والملف `deploy.mjs` يبنيه بالأمر @@ -51,4 +51,4 @@ curl -N http://:3000/chat \ المنفذ 3000 يعمل بـ HTTP عادي، فضع Traefik (المثبَّت في قالب الـ VPS) أو أي أداة أخرى لإنهاء TLS أمامه في أي استخدام حقيقي، وفق -[وثائق النشر](https://github.com/LinuxDevil/agent-sdk/blob/main/docs/deployment.md). +[وثائق النشر](https://github.com/LinuxDevil/lousho/blob/main/docs/deployment.md). diff --git a/ar/examples/phase-pipeline.mdx b/ar/examples/phase-pipeline.mdx index d680225..008dc2c 100644 --- a/ar/examples/phase-pipeline.mdx +++ b/ar/examples/phase-pipeline.mdx @@ -2,7 +2,7 @@ title: "خط المراحل (phase-pipeline)" sidebarTitle: "خط المراحل" --- -{/* Generated from examples/phase-pipeline/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/phase-pipeline/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} **هندسة المراحل** هي خط مراحل حتمي يملأ فيه النموذج كل مرحلة، بينما يعود *الترتيب* و*البوابة* وتعريف *الاكتمال* إلى الـ harness. هذا هو الشكل الذي تقوم عليه مستودعات مسارات العمل الوكيلية (phases-oss وleeovery/agentic-workflows): diff --git a/ar/examples/waves.mdx b/ar/examples/waves.mdx index 37f4abc..5b76a38 100644 --- a/ar/examples/waves.mdx +++ b/ar/examples/waves.mdx @@ -2,7 +2,7 @@ title: "الموجات (waves)" sidebarTitle: "الموجات" --- -{/* Generated from examples/waves/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/waves/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} **Waves** (Workers · Aggregate · Verify · Extend، أي العمّال والتجميع والتحقق والتوسيع) هو نمط التوازي المحدود المأخوذ من مسارات عمل "waves" المجتمعية (rayfernando-skills وWave-Engineering/claudecode-workflow). على خلاف التوزيع البسيط للمهام (fan-out)، تملك الموجة *عمودًا فقريًا للتحقق*: تُدمَج مخرجات العمّال بطريقة حتمية، ثم يحكم مُتحقِّق (verifier) على الناتج المجمَّع وفق مخطط مُنمَّط، ولا تعمل موجة لاحقة **إلا** على الثغرات التي يسمّيها المُتحقِّق. @@ -36,6 +36,6 @@ npm run example:waves # offline, scripted workers + verifier OPENROUTER_API_KEY=… npx tsx examples/waves/index.ts # live ``` -متى تلجأ إليه: في العمل الذي يتّسع عرضًا أولًا ويمكن فيه التحقق من *التغطية*، مثل تلخيص مجموعة نصوص أو تدقيق قاعدة كود أو ترحيل ملفات، وحين تكون الإجابة الجزئية أسوأ من الإجابة البطيئة. وحين يستطيع منسّق واحد أن يستوعب المهمة كلها، يكون شكل [deep-research](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/deep-research/)، الذي يوزّع فيه الـ agent الرئيسي المهام الفرعية، أبسط. +متى تلجأ إليه: في العمل الذي يتّسع عرضًا أولًا ويمكن فيه التحقق من *التغطية*، مثل تلخيص مجموعة نصوص أو تدقيق قاعدة كود أو ترحيل ملفات، وحين تكون الإجابة الجزئية أسوأ من الإجابة البطيئة. وحين يستطيع منسّق واحد أن يستوعب المهمة كلها، يكون شكل [deep-research](https://github.com/LinuxDevil/lousho/blob/main/examples/deep-research/)، الذي يوزّع فيه الـ agent الرئيسي المهام الفرعية، أبسط. راجع [docs/prompting-techniques.md](/ar/prompting-techniques) لمعرفة موقع هندسة الموجات في الفهرس الكامل للأنماط. diff --git a/ar/executor-api.mdx b/ar/executor-api.mdx index 56bf7d5..00c23a7 100644 --- a/ar/executor-api.mdx +++ b/ar/executor-api.mdx @@ -2,7 +2,7 @@ title: "Executor API" sidebarTitle: "Executor API" --- -{/* Generated from docs/executor-api.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/executor-api.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `createAgent()` هي نقطة الدخول الموصى بها (راجع [البدء السريع](/ar/quickstart)). هذه الصفحة للكود الذي يحتاج الخيارات الأدنى مستوى التي لا تأخذها `createAgent()`. ولنقل كود executor موجود إلى `createAgent()`، راجع [الانتقال إلى createAgent()](/ar/migrating-to-create-agent). diff --git a/ar/flows.mdx b/ar/flows.mdx index 994df6d..fbf41d1 100644 --- a/ar/flows.mdx +++ b/ar/flows.mdx @@ -1,7 +1,7 @@ --- title: "Flows" --- -{/* Generated from docs/flows.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/flows.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـflow هو سير عمل ثابت من عدة خطوات، يشغّله agent واحد. تكتب الخطوات وترتيبها مسبقاً، بدلًا من أن تترك القرار للـmodel. صف الـgraph بـ`FlowBuilder`، وشغّله بالدالة static `FlowExecutor.execute(flow, context, onEvent?)`. diff --git a/ar/guardrails.mdx b/ar/guardrails.mdx index 777d39a..9a936f0 100644 --- a/ar/guardrails.mdx +++ b/ar/guardrails.mdx @@ -1,7 +1,7 @@ --- title: "Guardrails و sandboxing" --- -{/* Generated from docs/guardrails.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/guardrails.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} أدوات تشغّل الـ agents بأمان: **input و output guardrails** تفحص ما يدخل أي run وما يخرج منه (ومعاملات الـ tool calls)، و**patch checks** فحوص fail-closed على تغيير مقترح قبل أن تتصرّف بناءً عليه، و**sandboxed tools** تعمل داخل container معزولة بدلًا من عملية الـ host. للقرارات البشرية على كل call انظر [Approvals](/ar/approvals)؛ وللـ hooks التي تفحص كل tool call أو ترفضه انظر [Hooks](/ar/hooks). @@ -112,7 +112,7 @@ if (result.guardrail?.info?.category === 'prompt-injection') console.warn(result ## Patch checks -`runPatchChecks(patch, checks)` يشغّل كل الفحوص بالتوازي على patch مقترح ويجمع النتائج في verdict واحد. [مثال ops-pipeline](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/ops-pipeline) يستخدمه كبوابة على patch من agent إصلاح قبل أن يُفتح pull request: +`runPatchChecks(patch, checks)` يشغّل كل الفحوص بالتوازي على patch مقترح ويجمع النتائج في verdict واحد. [مثال ops-pipeline](https://github.com/LinuxDevil/lousho/blob/main/examples/ops-pipeline) يستخدمه كبوابة على patch من agent إصلاح قبل أن يُفتح pull request: ```ts import { runPatchChecks, secretScanCheck, createDiffSizeCheck, createCommandCheck } from 'lousho'; diff --git a/ar/handoffs.mdx b/ar/handoffs.mdx index 5e02fcd..cbaeed9 100644 --- a/ar/handoffs.mdx +++ b/ar/handoffs.mdx @@ -2,7 +2,7 @@ title: "Handoffs" description: "الـ handoff يمرّر المحادثة كلها إلى agent آخر. الـ triage agent يقرأ طلب المستخدم ويسلّمه إلى متخصص. المتخصص يجيب المستخدم مباشرة، وفي الـ session (الجلسة) يبقى هو صاحب المحادثة في الدورات التالية." --- -{/* Generated from docs/handoffs.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/handoffs.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## handoffs أم sub-agents؟ diff --git a/ar/hooks.mdx b/ar/hooks.mdx index 7237a78..c4d8433 100644 --- a/ar/hooks.mdx +++ b/ar/hooks.mdx @@ -1,7 +1,7 @@ --- title: "Hooks" --- -{/* Generated from docs/hooks.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/hooks.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ hook هو object بالشكل `{ name, preToolCall?, postToolCall?, preGenerate?, postGenerate? }` يعمل داخل الـ agent loop، حول كل tool call وكل model call. استخدمه لكتابة audit log، أو لرفض tool، أو لحجب نتيجة أو قصّها، أو لحقن context قبل الـ model call، أو لتغيير الـ request الذي سيُرسَل (header أو `temperature`). diff --git a/ar/hosted-tools.mdx b/ar/hosted-tools.mdx index e3bbc7e..f9a312d 100644 --- a/ar/hosted-tools.mdx +++ b/ar/hosted-tools.mdx @@ -1,7 +1,7 @@ --- title: "Hosted tools عند الـ provider" --- -{/* Generated from docs/hosted-tools.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/hosted-tools.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} بعض الـ providers يشغّلون tools بنفسهم، داخل طلب الـ model نفسه. أشهرها عند OpenAI: web search وcode interpreter وfile search. ضعها في `tools` بجانب الـ tools التي تعرّفها أنت. الـ provider هو الذي يشغّلها، والـ run يبلّغ عن كل call في الـ events، ويحتفظ بها في الـ transcript (سجل الرسائل)، ويحسبها في الـ usage. diff --git a/ar/hostinger-vps.mdx b/ar/hostinger-vps.mdx index cda91ad..00a516e 100644 --- a/ar/hostinger-vps.mdx +++ b/ar/hostinger-vps.mdx @@ -1,11 +1,11 @@ --- title: "النشر على Hostinger VPS" --- -{/* Generated from docs/hostinger-vps.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/hostinger-vps.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ target `docker` في [النشر](/ar/deployment) هو Dockerfile عادي مع سيرفر مُجمَّع (bundled)، لذلك يعمل على أي host فيه Docker، ومنها Hostinger VPS. -والمثال [hostinger-deploy](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/hostinger-deploy) +والمثال [hostinger-deploy](https://github.com/LinuxDevil/lousho/tree/main/examples/hostinger-deploy) يقدّم وصفة كاملة ومُختبَرة: `deploy.mjs` يتحقق من الـ VM عبر Hostinger API، ويبني محليًا، وينقل الـ context عبر SSH، ويشغّل الحاوية، ثم يثبت أن النشر نجح عبر `/health` ودورة `/chat` حقيقية واحدة. diff --git a/ar/improve.mdx b/ar/improve.mdx index cba860b..2391c43 100644 --- a/ar/improve.mdx +++ b/ar/improve.mdx @@ -2,7 +2,7 @@ title: "تحسين الـ agent" description: "تبحث improveAgent عن harness أفضل دون اتصال، وتختبره مقابل الحالي، وتُرجع اقتراحًا يراجعه شخص." --- -{/* Generated from docs/improve.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/improve.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} تبحث `improveAgent()` عن مجموعة أفضل من التعليمات والمهارات (skills) للـ agent، مستخدمةً حالات فاشلة تعطيها لها. هي تُرجع اقتراحًا فقط: لا يُطبَّق شيء، diff --git a/ar/installation.mdx b/ar/installation.mdx index 5b8ad00..7999e9f 100644 --- a/ar/installation.mdx +++ b/ar/installation.mdx @@ -1,7 +1,7 @@ --- title: "التثبيت" --- -{/* Generated from docs/installation.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/installation.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## المتطلبات @@ -150,7 +150,7 @@ npm install ./lousho-.tgz ai zod على Windows، الأمر `npx C:\abs\path\create-lousho-.tgz` لا يفعل شيئًا بصمت (سلوك غريب في npm مع المسارات المطلقة التي تبدأ بحرف قرص). استخدم بدلًا منه مسارًا نسبيًّا (`npx ./create-lousho-.tgz my-agent`) أو صيغة `file:` (`npx file:C:/abs/path/create-lousho-.tgz my-agent`). -الأمر `npm install github:LinuxDevil/agent-sdk` **لا** يعمل: `dist/` غير موجود في git، وليس في المستودع خطوة `prepare` تعمل build، فينتهي التثبيت بلا نقاط دخول. +الأمر `npm install github:LinuxDevil/lousho` **لا** يعمل: `dist/` غير موجود في git، وليس في المستودع خطوة `prepare` تعمل build، فينتهي التثبيت بلا نقاط دخول. ## إنشاء مشروع جديد diff --git a/ar/mcp-integration.mdx b/ar/mcp-integration.mdx index 93c8f56..1d84303 100644 --- a/ar/mcp-integration.mdx +++ b/ar/mcp-integration.mdx @@ -2,7 +2,7 @@ title: "MCP (Model Context Protocol)" sidebarTitle: "MCP" --- -{/* Generated from docs/mcp.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/mcp.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} MCP يعمل في الاتجاهين. الـ agent يستطيع استدعاء tools موجودة في سيرفرات MCP، أو يكون هو نفسه سيرفر MCP يناديه Claude Code وCursor وغيرهما من MCP clients. diff --git a/ar/memory.mdx b/ar/memory.mdx index e7a7854..c4fe4e3 100644 --- a/ar/memory.mdx +++ b/ar/memory.mdx @@ -1,7 +1,7 @@ --- title: "الذاكرة" --- -{/* Generated from docs/memory.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/memory.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ [session](/ar/sessions) (الجلسة) تتذكّر محادثة واحدة. أما الذاكرة فهي ما يحتفظ به الـ agent **عبر** المحادثات: تفضيلات المستخدم، وحقائق قيلت له، وقرارات اتُّخذت الأسبوع الماضي. تعطي الـ agent واحدًا أو أكثر من **memory slots** بأسماء؛ يخزّن كل slot عناصر نصية قصيرة عن طريق provider، ومفتاحها **scope**: ذاكرة واحدة للجميع، أو واحدة لكل session، أو واحدة لكل مستخدم. diff --git a/ar/migrating-from-ai-sdk.mdx b/ar/migrating-from-ai-sdk.mdx index 8e6e4b3..c0272e8 100644 --- a/ar/migrating-from-ai-sdk.mdx +++ b/ar/migrating-from-ai-sdk.mdx @@ -2,7 +2,7 @@ title: "الانتقال من Vercel AI SDK" sidebarTitle: "الانتقال من الـ AI SDK" --- -{/* Generated from docs/migrating-from-ai-sdk.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/migrating-from-ai-sdk.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} هذه الصفحة للكود المبني على حلقة الـ agent في Vercel AI SDK: `generateText({ tools })` أو `streamText({ tools })` مع حد للخطوات، أو `new ToolLoopAgent({ ... })`. يمكنك أن تحتفظ diff --git a/ar/migrating-to-create-agent.mdx b/ar/migrating-to-create-agent.mdx index e667367..0e35a73 100644 --- a/ar/migrating-to-create-agent.mdx +++ b/ar/migrating-to-create-agent.mdx @@ -2,7 +2,7 @@ title: "الانتقال إلى createAgent()" sidebarTitle: "الانتقال إلى createAgent()" --- -{/* Generated from docs/migrating-to-create-agent.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/migrating-to-create-agent.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} هذه الصفحة للكود الذي يستخدم `AgentBuilder` و`AgentExecutor` و`ToolRegistry` و`resumeAfterApproval()`: الـ API الأدنى مستوى الذي بدأ به الـ SDK. هذه الـ exports انتقلت إلى `lousho/executor` ولن تُحذف. إذا بقيت عليها، راجع [واجهة الـ executor](/ar/executor-api). سبب الانتقال أن استدعاء `createAgent()` واحد يربط الـ tools والـ stores والـ approvals (موافقات) والـ sessions (جلسات) والـ retries (إعادة المحاولة) والـ compaction (ضغط سياق المحادثة)، بينما الـ executor يجعلك تجمع كل واحد منها بنفسك وتمرره في كل استدعاء. diff --git a/ar/models-and-cost.mdx b/ar/models-and-cost.mdx index 4d7f889..93458dc 100644 --- a/ar/models-and-cost.mdx +++ b/ar/models-and-cost.mdx @@ -2,7 +2,7 @@ title: "الـ models والـ tokens والتكلفة" sidebarTitle: "الـ models والتكلفة" --- -{/* Generated from docs/models-and-cost.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/models-and-cost.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## الـ models وجدول الأسعار diff --git a/ar/nextjs.mdx b/ar/nextjs.mdx index c5766e6..54cbd29 100644 --- a/ar/nextjs.mdx +++ b/ar/nextjs.mdx @@ -2,7 +2,7 @@ title: "Next.js وframeworks أخرى على Fetch API" sidebarTitle: "Next.js وFetch frameworks" --- -{/* Generated from docs/nextjs.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/nextjs.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `createRouteHandler(agent, options?)` يشغّل الـ agent من ملف route في أي framework تستقبل الـ handlers فيه `Request` من Web API وترجع `Response`: الـ Next.js App Router، وSvelteKit، وRemix / React Router، وHono، وBun.serve. هو غلاف رفيع فوق نفس session API المستخدم في `lousho dev` وفي السيرفر المنشور ([النشر](/ar/deployment))، لذلك تعمل الـ bindings الخاصة بـ React وVue وSvelte معه بدون أي تغيير. لا يستورد أي framework ولا أي module من `node:*`. بادئة `vendor/` في اسم الـ model هي التي تحدد الـ provider؛ ومع OpenRouter استخدم `openrouter//` (مثل `openrouter/openai/gpt-4o-mini`). diff --git a/ar/oauth.mdx b/ar/oauth.mdx index d0a583c..1ef4df9 100644 --- a/ar/oauth.mdx +++ b/ar/oauth.mdx @@ -1,7 +1,7 @@ --- title: "OAuth" --- -{/* Generated from docs/oauth.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/oauth.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الأدوات وخوادم MCP التي تستدعي API بالنيابة عن أحد تحتاج access token من OAuth، وعادةً refresh token لكي تجلب الذي يليه. هذه الصفحة تغطي أين تعيش هذه الـtokens: في جزء `tokens` من `AgentStore`، مفهرسةً بالـprovider وبمالك الـcredential. وتغطي كيف تحصل الأداة على واحد: تستدعي `ctx.getToken(provider)`، وإذا لم يربط المستخدم حسابه بعد يتوقف الـrun إلى أن يسجّل الدخول ([أدوات تحتاج تسجيل دخول](#أدوات-تحتاج-تسجيل-دخول)). وخوادم MCP عبر HTTP تسجّل الدخول بـOAuth أيضًا ([خوادم MCP مع OAuth](#خوادم-mcp-مع-oauth)). diff --git a/ar/observability.mdx b/ar/observability.mdx index 827de69..c8fad6e 100644 --- a/ar/observability.mdx +++ b/ar/observability.mdx @@ -2,7 +2,7 @@ title: "Tracing و observability" sidebarTitle: "Tracing" --- -{/* Generated from docs/observability.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/observability.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `AgentExecutor.execute()` و`FlowExecutor.execute()` ينتجان spans عبر `TraceExporter`. أسماء الـ spans والـ attributes تتبع [GenAI semantic conventions في OpenTelemetry](https://opentelemetry.io/docs/specs/semconv/gen-ai/)، فالـ observability backends التي تفهم GenAI تتعرف عليها بدون mapping خاص. diff --git a/ar/openapi-tools.mdx b/ar/openapi-tools.mdx index 8fb19be..abb8b81 100644 --- a/ar/openapi-tools.mdx +++ b/ar/openapi-tools.mdx @@ -1,7 +1,7 @@ --- title: "OpenAPI tools" --- -{/* Generated from docs/openapi-tools.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/openapi-tools.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} أغلب الخدمات تنشر مستند OpenAPI وما عندها سيرفر MCP. `openApiTools()` يحوّل المستند إلى tool واحد typed لكل عملية في الـ API، فيقدر الـ agent diff --git a/ar/pause-and-resume.mdx b/ar/pause-and-resume.mdx index 04e68f5..805a292 100644 --- a/ar/pause-and-resume.mdx +++ b/ar/pause-and-resume.mdx @@ -2,7 +2,7 @@ title: "الإيقاف المؤقت وإعادة التشغيل والاستئناف" description: "درس عملي لمهمة واحدة: agent يتوقف ليسأل إنسانًا قبل استدعاء أداة خطِر، ويصمد أمام إعادة تشغيل الـ process وهو ينتظر، ويكمل حين يقرّر أحدهم. ثم ما يحدث حين يموت الـ process في منتصف أداة." --- -{/* Generated from docs/pause-and-resume.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/pause-and-resume.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} تستخدم المقتطفات `mockModel` لكي تعمل كما تُلصَق، دون مفتاح API. ضع `provider` أو `model` حقيقيًا للـ agent الخاص بك؛ ولا شيء آخر يتغيّر. كل diff --git a/ar/permission-modes.mdx b/ar/permission-modes.mdx index f38a174..56f78c5 100644 --- a/ar/permission-modes.mdx +++ b/ar/permission-modes.mdx @@ -1,7 +1,7 @@ --- title: "أوضاع الصلاحيات" --- -{/* Generated from docs/permission-modes.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/permission-modes.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} وضع الصلاحيات يضع الـ agent في حالة لها اسم، بدلًا من أن تكتب rules: ينظر فقط دون أن يلمس شيئًا (`plan`)، أو يعدّل الملفات دون سؤال (`acceptEdits`)، أو لا يتوقف أبدًا (`dontAsk`). بدّله في منتصف الـ session، بالطريقة نفسها التي تعمل بها مع coding agent: خطّط أولًا، ثم دعه يعدّل. diff --git a/ar/prompting-techniques.mdx b/ar/prompting-techniques.mdx index ad26d8c..4e0807a 100644 --- a/ar/prompting-techniques.mdx +++ b/ar/prompting-techniques.mdx @@ -2,7 +2,7 @@ title: "تقنيات الـ prompting وإدارة السياق" description: "جودة الـ agent تأتي من ثلاث طبقات، ومن المفيد أن تعرف في أي طبقة تقع كل تقنية، لأن إصلاح agent يتصرف بشكل خاطئ يختلف من طبقة لأخرى:" --- -{/* Generated from docs/prompting-techniques.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/prompting-techniques.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} 1. **تقنيات على مستوى الـ prompt**: ما يُطلب من استدعاء model واحد، وكيف يُصاغ هذا الطلب. @@ -26,11 +26,11 @@ Lousho يعطيك الطبقتين الثانية والثالثة كآليات | Few-shot (التعلّم داخل السياق) | تعرض أزواج مدخلات/مخرجات فينسخ الـ model شكلها | ضع أزواج الأمثلة في `instructions`، أو جهّز دورات مسبقة عبر `agent.session().send()`: تبادل "user/assistant" مكتوب مسبقًا تكمله الدورة الحقيقية | | Chain-of-thought | "فكّر خطوة بخطوة" قبل الإجابة | تعليمات. والأفضل استخدام جهد `reasoning` حيث يدعمه الـ provider: tokens الاستدلال أنظف من مونولوج مطلوب في الـ prompt | | Self-consistency | تأخذ N إجابة وتختار رأي الأغلبية | N run مع خطوة تصويت (توزيع `parallel`/`llmCall` في flow، أو sub-agent لكل عيّنة) | -| Tree/graph-of-thoughts | تستكشف بحثًا متفرعًا في مسارات الاستدلال | sub-agent لكل فرع، و aggregator يقارن بينها: نفس شكل التوزيع في [deep-research](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/deep-research) | +| Tree/graph-of-thoughts | تستكشف بحثًا متفرعًا في مسارات الاستدلال | sub-agent لكل فرع، و aggregator يقارن بينها: نفس شكل التوزيع في [deep-research](https://github.com/LinuxDevil/lousho/tree/main/examples/deep-research) | | ReAct | تبادل بين الاستدلال والتنفيذ | حلقة الـ agent نفسها: كل حلقة tools في Lousho هي ReAct | -| Reflexion | تنتقد المحاولة السابقة، ثم تعيد المحاولة مع هذا النقد | [evaluator-loop](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/evaluator-loop): ملاحظات `llmCritique` تُنقل إلى المسودة التالية | +| Reflexion | تنتقد المحاولة السابقة، ثم تعيد المحاولة مع هذا النقد | [evaluator-loop](https://github.com/LinuxDevil/lousho/tree/main/examples/evaluator-loop): ملاحظات `llmCritique` تُنقل إلى المسودة التالية | | Least-to-most | تحل المسألة الفرعية الأسهل أولًا، ثم تتدرج إلى الأصعب | تعليمات تقسّم المهمة إلى خطوات مرتبة، أو تسلسل `flows` | -| Plan-and-solve / plan-and-execute | تكتب خطة، ثم تنفذها | `permissionMode: 'plan'`: يجب أن يقدّم الـ agent خطة قبل تشغيل أي tool كتابة؛ راجع [plan-mode](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/plan-mode) | +| Plan-and-solve / plan-and-execute | تكتب خطة، ثم تنفذها | `permissionMode: 'plan'`: يجب أن يقدّم الـ agent خطة قبل تشغيل أي tool كتابة؛ راجع [plan-mode](https://github.com/LinuxDevil/lousho/tree/main/examples/plan-mode) | | Skeleton-of-thought | مخطط أولًا، ثم توسيع كل نقطة (بالتوازي) | توزيع `task` على نقاط المخطط: شكل deep-research | | Self-ask | الـ model يسأل أسئلته اللاحقة ويجيب عنها بنفسه | تعليمات، أو flow صريح بـ `oneOf` يكرر sub-agent للتوضيح | | Generated knowledge | تطلب من الـ model ذكر الحقائق المتعلقة، ثم يجيب | `llmCall` أول يكتب في متغير، والـ prompt الثاني يُبنى من قالب يستخدمه | @@ -91,14 +91,14 @@ tokens كثيرة (نتائج البحث الكاملة، وقراءة المل | النمط | الشكل | المثال في Lousho | | ------- | ----- | -------------- | | Prompt chaining | خطوة ← خطوة ← خطوة، كل واحدة استدعاء | تسلسلات `flows` (`llmCall` ← `llmCall`) | -| Routing | صنّف، ثم وزّع على متخصص | [workflow-router](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/workflow-router)؛ والنسخة المتينة: handoffs في [support-desk](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/support-desk) | +| Routing | صنّف، ثم وزّع على متخصص | [workflow-router](https://github.com/LinuxDevil/lousho/tree/main/examples/workflow-router)؛ والنسخة المتينة: handoffs في [support-desk](https://github.com/LinuxDevil/lousho/tree/main/examples/support-desk) | | Parallelization | N استدعاء في نفس الوقت، ثم دمج النتائج | استدعاءات `task` متوازية (`toolConcurrency` غير محدود افتراضيًا) | -| Orchestrator-workers | agent رئيسي يقسّم العمل، و workers ينفذونه | [deep-research](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/deep-research) | -| Evaluator-optimizer | حلقة مسودة ← تقييم ← مراجعة | [evaluator-loop](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/evaluator-loop) (`llmCritique`) | -| **Wave engineering** | موجات محدودة من workers متوازية مع بوابة تحقق وتوسيع محدود النطاق (WAVES: Workers · Aggregate · Verify · Extend) | [waves](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/waves) | -| **Phase engineering** | خطوط مراحل حتمية (spec ← plan ← implement ← verify) لا يستطيع الـ model تخطيها، مع بوابات أوامر إثبات | [phase-pipeline](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/phase-pipeline) (flows مع خطوات إثبات حقيقية) | +| Orchestrator-workers | agent رئيسي يقسّم العمل، و workers ينفذونه | [deep-research](https://github.com/LinuxDevil/lousho/tree/main/examples/deep-research) | +| Evaluator-optimizer | حلقة مسودة ← تقييم ← مراجعة | [evaluator-loop](https://github.com/LinuxDevil/lousho/tree/main/examples/evaluator-loop) (`llmCritique`) | +| **Wave engineering** | موجات محدودة من workers متوازية مع بوابة تحقق وتوسيع محدود النطاق (WAVES: Workers · Aggregate · Verify · Extend) | [waves](https://github.com/LinuxDevil/lousho/tree/main/examples/waves) | +| **Phase engineering** | خطوط مراحل حتمية (spec ← plan ← implement ← verify) لا يستطيع الـ model تخطيها، مع بوابات أوامر إثبات | [phase-pipeline](https://github.com/LinuxDevil/lousho/tree/main/examples/phase-pipeline) (flows مع خطوات إثبات حقيقية) | | Autonomous loop | حلقة tools مفتوحة حتى الانتهاء | حلقة الـ agent الافتراضية، محدودة بـ `maxSteps` و `limits` والصلاحيات | -| Handoffs | التحكم ينتقل بالكامل إلى متخصص | [support-desk](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/support-desk)، و [handoffs](/ar/handoffs) | +| Handoffs | التحكم ينتقل بالكامل إلى متخصص | [support-desk](https://github.com/LinuxDevil/lousho/tree/main/examples/support-desk)، و [handoffs](/ar/handoffs) | ### قرارات typed (بأسلوب "System One") @@ -121,7 +121,7 @@ const triage = createAgent({ // and confidence < 0.7 routes to a human queue rather than acting. ``` -[coding-agent-workflows](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/coding-agent-workflows) +[coding-agent-workflows](https://github.com/LinuxDevil/lousho/tree/main/examples/coding-agent-workflows) هو المثال الكامل: coding agent يقوده سير عمل حتمي نقاط تفرّعه قرارات typed، وحين تكون الثقة تحت الحد الأدنى يخرج بـ `needs-human` بدل تطبيق patch. diff --git a/ar/providers.mdx b/ar/providers.mdx index 089b0d6..5371e34 100644 --- a/ar/providers.mdx +++ b/ar/providers.mdx @@ -1,7 +1,7 @@ --- title: "Providers" --- -{/* Generated from docs/providers.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/providers.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ `LLMProvider` هو ما يولّد الـ agent من خلاله. يأتي الـ SDK مع providers لـ OpenAI وAnthropic وGoogle (Gemini) وOpenRouter وOllama و`pi` (لكلٍّ منها @@ -227,7 +227,7 @@ console.log(result.text); | ------ | ----------- | | `resolveProvider('p/model')` | يبني provider حقيقيًا من credentials البيئة. | | `LLMProviderRegistry` | registry يحتوي على factories للـ providers: `create(type, config)`، `register(type, factory)`، `has(type)`. | -| `OpenAIProvider`, `AnthropicProvider`, `GoogleProvider`, `OllamaProvider`, `OpenRouterProvider`, `PiProvider` | كلاسات الـ providers. راجع [examples/openrouter](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/openrouter) لميزات OpenRouter الخاصة. `PiProvider` هو الـ adapter الخاص بـ `pi/...` أدناه. | +| `OpenAIProvider`, `AnthropicProvider`, `GoogleProvider`, `OllamaProvider`, `OpenRouterProvider`, `PiProvider` | كلاسات الـ providers. راجع [examples/openrouter](https://github.com/LinuxDevil/lousho/blob/main/examples/openrouter) لميزات OpenRouter الخاصة. `PiProvider` هو الـ adapter الخاص بـ `pi/...` أدناه. | | `createMockProvider()`, `MockLLMProvider` | ردود جاهزة للعروض التجريبية؛ يحاكي tool call عندما تسمّي الرسالة tool. | | `mockModel(script)` | provider يعمل بسكربت للاختبارات ويتحقق من التسلسل (من `/testing`)؛ راجع [Testing](/ar/testing). | | `withRetry()`, `withFallback()` | يعيد محاولة الفشل المؤقت مع backoff وينتقل إلى الـ provider التالي؛ راجع [Provider retries وfallback](/ar/configuration#الـ-retry-والـ-fallback-عند-الـ-provider). | diff --git a/ar/queue-and-steer.mdx b/ar/queue-and-steer.mdx index fec3e08..0152fcc 100644 --- a/ar/queue-and-steer.mdx +++ b/ar/queue-and-steer.mdx @@ -2,7 +2,7 @@ title: "Queued input and steering" sidebarTitle: "Queued input and steering" --- -{/* Generated from docs/queue-and-steer.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/queue-and-steer.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} أضف input إلى run ما زال يعمل، أو غيّر اتجاهه. كلاهما methods على الـ run handle الذي تُرجعه `agent.stream()`؛ راجع [Streaming](/ar/streaming). diff --git a/ar/quickstart.mdx b/ar/quickstart.mdx index f9e37c6..dd92492 100644 --- a/ar/quickstart.mdx +++ b/ar/quickstart.mdx @@ -3,7 +3,7 @@ title: "البدء السريع" sidebarTitle: "البدء السريع" description: "هذه الصفحة توصلك من agent بسطر واحد إلى tools وstreaming وsessions وapproval واختبار offline. كل مثال بعد الأول يعمل كما هو دون مفتاح API، على الـ mock provider المدمج في الـ SDK." --- -{/* Generated from docs/quick-start.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/quick-start.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## ابدأ مشروع جديد @@ -199,7 +199,7 @@ console.log('ok'); ## 7. Custom provider -`model: 'provider/model'` هو الطريق المختصر. مرّر `provider` بدلًا منه إذا كان عندك `LLMProvider` خاص بك أو تريد الـ mock للاختبارات. `resolveProvider('/')` هي الدالة التي يستخدمها `model` من الداخل. هذا المثال يستخدم OpenAI إذا كان `OPENAI_API_KEY` موجودًا، ويرجع إلى الـ mock provider في غير ذلك. وهو نفس النمط الذي تستخدمه [الأمثلة](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/README.md) الجاهزة للتشغيل. +`model: 'provider/model'` هو الطريق المختصر. مرّر `provider` بدلًا منه إذا كان عندك `LLMProvider` خاص بك أو تريد الـ mock للاختبارات. `resolveProvider('/')` هي الدالة التي يستخدمها `model` من الداخل. هذا المثال يستخدم OpenAI إذا كان `OPENAI_API_KEY` موجودًا، ويرجع إلى الـ mock provider في غير ذلك. وهو نفس النمط الذي تستخدمه [الأمثلة](https://github.com/LinuxDevil/lousho/blob/main/examples/README.md) الجاهزة للتشغيل. ```ts import { createAgent, createMockProvider, resolveProvider } from 'lousho'; @@ -265,4 +265,4 @@ npx lousho build --target=node-server --agent=agent.yaml - [الإعداد](/ar/configuration) - كل حقول ملف الـ spec، ومتغيرات بيئة الـ providers، وflags الـ CLI. - [النشر](/ar/deployment) - أهداف `lousho build` (سيرفر Node، وDocker، وCloudflare Workers). - [نظرة عامة على الـ API](/ar/api-overview) - أهم الـ exports ومكان مرجع الـ API الكامل. -- [الأمثلة](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/README.md) - agents جاهزة للتشغيل. +- [الأمثلة](https://github.com/LinuxDevil/lousho/blob/main/examples/README.md) - agents جاهزة للتشغيل. diff --git a/ar/react.mdx b/ar/react.mdx index 71b4907..1fafad7 100644 --- a/ar/react.mdx +++ b/ar/react.mdx @@ -1,7 +1,7 @@ --- title: "React" --- -{/* Generated from docs/react.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/react.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `useLoushoAgent()` يضع agent خلف واجهة محادثة. يشغّل turn، ويقرأ الـ [typed event stream](/ar/streaming) فور وصوله، ويحوّله إلى state جاهزة للـ render: رسائل بنصها وtool calls، و`status` لحقل كتابة الرسالة، والـ tool call المتوقف على approval إن وُجد. diff --git a/ar/reasoning.mdx b/ar/reasoning.mdx index 44194c1..e43ee99 100644 --- a/ar/reasoning.mdx +++ b/ar/reasoning.mdx @@ -2,7 +2,7 @@ title: "Reasoning" --- -{/* Generated from docs/reasoning.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/reasoning.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} موديلات الـ reasoning تفكّر قبل أن ترد. خيار `reasoning` يحدد مقدار التفكير. والـ run يُرجع لك ما فكّر فيه الموديل منفصلًا عن الجواب: كأحداث `reasoning.*` diff --git a/ar/registry.mdx b/ar/registry.mdx index b6190e5..689604e 100644 --- a/ar/registry.mdx +++ b/ar/registry.mdx @@ -1,7 +1,7 @@ --- title: "Registry" --- -{/* Generated from docs/registry.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/registry.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `lousho add` يثبّت tool أو skill أو channel أو schedule أو خانة memory أو kit داخل [agent directory](/ar/agent-directories) من registry. عنصر الـ`kit` هو @@ -16,7 +16,7 @@ agent directory كامل: تثبيته يوزّع `agent.*` و`instructions.md` `registry/dist/` في مستودع الـSDK مُقدَّم كـJSON ثابت. إذا تعذّر الوصول لهذا الـhost، يقرأ `lousho add` بدلًا منه من GitHub: من وسم الإصدار (release tag) الخاص بنسخة الـSDK المثبّتة -(`https://raw.githubusercontent.com/LinuxDevil/agent-sdk//registry/dist/index.json`)، +(`https://raw.githubusercontent.com/LinuxDevil/lousho//registry/dist/index.json`)، ثم من الوسم نفسه مع البادئة `v` (للإصدارات الأقدم)، ثم من `main`. وإذا تعذّر الوصول إلى GitHub أيضًا (بدون اتصال، أو عطل، أو نسخة لم يُرفع وسمها أبدًا)، يقرأ `lousho add` نسخة `registry/dist/` المرفقة داخل الحزمة المثبّتة، diff --git a/ar/runs.mdx b/ar/runs.mdx index b8a5641..82970a5 100644 --- a/ar/runs.mdx +++ b/ar/runs.mdx @@ -3,7 +3,7 @@ title: "Runs: أسباب الانتهاء والإلغاء والـtools الم sidebarTitle: "Runs والإلغاء" description: "كيف ينتهي الـrun، كيف توقفه، وكيف تعمل الـtool calls في دورة model واحدة بالتوازي." --- -{/* Generated from docs/runs.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/runs.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## أسباب الانتهاء diff --git a/ar/schedules.mdx b/ar/schedules.mdx index 4392e82..d2c7f73 100644 --- a/ar/schedules.mdx +++ b/ar/schedules.mdx @@ -1,7 +1,7 @@ --- title: "Schedules" --- -{/* Generated from docs/schedules.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/schedules.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ **schedule** يشغّل الـ agent على تعبير cron، بدون أن يطلبه أحد: ملخص صباحي، تنظيف ليلي. تعرّف واحدًا بـ `defineSchedule()`، و[مجلد الـ agent](/ar/agent-directories) يلتقطها من `schedules/`، و`startSchedules()` (أو سيرفر node) يشغّلها داخل العملية. الـ schedules حلّت محل `CronTriggerAdapter` (راجع [الانتقال من الـ triggers](/ar/upgrading#الانتقال-من-الـ-triggers)). diff --git a/ar/self-healing-paper.mdx b/ar/self-healing-paper.mdx index f2ca3b9..7820857 100644 --- a/ar/self-healing-paper.mdx +++ b/ar/self-healing-paper.mdx @@ -4,7 +4,7 @@ description: "تقرير بأسلوب الأوراق البحثية عن حلق --- -هذه الصفحة تقرير، وليست دليل استخدام. تصف تصميم حلقة التعافي الذاتي في الـ SDK المسمّى `lousho`، والأدلة المجموعة حتى 2026-11-09. ولمعرفة طريقة الاستخدام، اقرأ [التعافي الذاتي والتحسين الذاتي](/ar/self-healing) و[تحسين الـ agent](/ar/improve). والملف [docs/improve.md](https://github.com/LinuxDevil/agent-sdk/blob/main/docs/improve.md) في مستودع الـ SDK هو المرجع الكامل للحلقة. +هذه الصفحة تقرير، وليست دليل استخدام. تصف تصميم حلقة التعافي الذاتي في الـ SDK المسمّى `lousho`، والأدلة المجموعة حتى 2026-11-09. ولمعرفة طريقة الاستخدام، اقرأ [التعافي الذاتي والتحسين الذاتي](/ar/self-healing) و[تحسين الـ agent](/ar/improve). والملف [docs/improve.md](https://github.com/LinuxDevil/lousho/blob/main/docs/improve.md) في مستودع الـ SDK هو المرجع الكامل للحلقة. ## الملخّص @@ -264,4 +264,4 @@ npx tsx examples/heal-benchmark/loop.ts --live --scenarios all --out ./m4-proof. npx tsx examples/heal-benchmark/loop.ts --report --out ./m4-proof.jsonl ``` -يطبع التقرير كل معيار مع أعداده وحكمًا واحدًا: GO، أو GO with misses، أو NO-GO، أو INCOMPLETE. وراجع [README](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/heal-benchmark/README.md) الخاص بالـ benchmark لمعرفة كل الخيارات. +يطبع التقرير كل معيار مع أعداده وحكمًا واحدًا: GO، أو GO with misses، أو NO-GO، أو INCOMPLETE. وراجع [README](https://github.com/LinuxDevil/lousho/blob/main/examples/heal-benchmark/README.md) الخاص بالـ benchmark لمعرفة كل الخيارات. diff --git a/ar/self-healing.mdx b/ar/self-healing.mdx index 6386b68..87db340 100644 --- a/ar/self-healing.mdx +++ b/ar/self-healing.mdx @@ -2,7 +2,7 @@ title: "التعافي الذاتي والتحسين الذاتي" description: "كيف يستعيد Lousho الـ run الفاشل الآن (agent.recover()، وإعادة محاولة مع تشخيص، والتوقف المؤقت أثناء انقطاع المزوّد، وحارس التقدّم) ويقترح تعديلات على التعليمات والمهارات لاحقًا (improveAgent)، يوافق عليها شخص دائمًا." --- -{/* Generated from docs/self-healing.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/self-healing.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} يتعامل Lousho مع حالات الفشل في طبقتين: diff --git a/ar/sessions.mdx b/ar/sessions.mdx index c7a14bf..054a7e2 100644 --- a/ar/sessions.mdx +++ b/ar/sessions.mdx @@ -1,7 +1,7 @@ --- title: "Sessions" --- -{/* Generated from docs/sessions.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/sessions.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `agent.send(text)` هو turn واحد فقط: كل call يبدأ من سجل فارغ. أما الـ **session** (جلسة) فهي محادثة من أكثر من turn: تحتفظ بسجل المحادثة diff --git a/ar/skills.mdx b/ar/skills.mdx index c96cdbb..3458e4b 100644 --- a/ar/skills.mdx +++ b/ar/skills.mdx @@ -2,7 +2,7 @@ title: "Skills" --- -{/* Generated from docs/skills.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/skills.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ skill هي مجموعة تعليمات لها اسم (مستند markdown) يحمّلها الـ model **فقط عندما يحتاجها**. هذه هي فكرة progressive disclosure: بدلًا من لصق كل playbook في الـ system prompt، تسرد الـ skills بالاسم مع وصف من سطر واحد، ويجلب الـ model النص الكامل عند الطلب. diff --git a/ar/stream-events.mdx b/ar/stream-events.mdx index 54a2d26..340f98d 100644 --- a/ar/stream-events.mdx +++ b/ar/stream-events.mdx @@ -2,7 +2,7 @@ title: "Stream events" sidebarTitle: "Stream events" --- -{/* Generated from docs/stream-events.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/stream-events.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الأحداث التي يطلقها `agent.stream()`، وترتيبها، وطريقة ترقيم إصداراتها. إذا أردت معرفة كيفية استهلاك الـ stream، فراجع [Streaming](/ar/streaming). diff --git a/ar/streaming.mdx b/ar/streaming.mdx index da3181c..3e8cac7 100644 --- a/ar/streaming.mdx +++ b/ar/streaming.mdx @@ -1,7 +1,7 @@ --- title: "Streaming" --- -{/* Generated from docs/streaming.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/streaming.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `agent.stream()` تشغّل الـ agent بطريقة `agent.send()` نفسها بالضبط، وتُرجع لك الـ run كـ stream من أحداث typed: نص الـ model حالما يصل، وكل tool call، diff --git a/ar/structured-output.mdx b/ar/structured-output.mdx index 5aff37a..a5238ce 100644 --- a/ar/structured-output.mdx +++ b/ar/structured-output.mdx @@ -1,7 +1,7 @@ --- title: "Structured output" --- -{/* Generated from docs/structured-output.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/structured-output.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} مرّر للـ agent zod schema في الخيار `output`، فيصبح الرد النهائي object typed وصالحًا وفق الـ schema: `result.object`. يستطيع الـ agent إجراء tool calls أولًا كالمعتاد؛ والإجابة النهائية وحدها يجب أن تطابق الـ schema. بادئة `vendor/` في اسم الـ model هي التي تحدد الـ provider؛ ومع OpenRouter استخدم `openrouter//` (مثال: `openrouter/openai/gpt-4o-mini`). diff --git a/ar/sub-agents.mdx b/ar/sub-agents.mdx index ad39cd5..befbeab 100644 --- a/ar/sub-agents.mdx +++ b/ar/sub-agents.mdx @@ -1,7 +1,7 @@ --- title: "Sub-agents" --- -{/* Generated from docs/sub-agents.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/sub-agents.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـsub-agent (وكيل فرعي) هو agent مستقل — له تعليمات وmodel وtools وskills خاصة به — يسلّمه الـlead (الوكيل الرئيسي) مهمة مكتفية بذاتها. diff --git a/ar/svelte.mdx b/ar/svelte.mdx index 963e5cd..91a55fd 100644 --- a/ar/svelte.mdx +++ b/ar/svelte.mdx @@ -2,7 +2,7 @@ title: "Svelte" --- -{/* Generated from docs/svelte.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/svelte.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `loushoAgent()` لـSvelte هي نفس [React hook](/ar/react) لكن بصيغة store: نفس المصادر، نفس الحالة، نفس الأوامر، فوق نفس [typed event stream](/ar/streaming). diff --git a/ar/testing.mdx b/ar/testing.mdx index 4fcc5ba..d6d52b9 100644 --- a/ar/testing.mdx +++ b/ar/testing.mdx @@ -1,7 +1,7 @@ --- title: "اختبار الـ agents" --- -{/* Generated from docs/testing.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/testing.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `lousho/testing` يوفّر `mockModel`: ‏`LLMProvider` يعمل بسكربت deterministic للـ unit tests. تكتب ما يُفترض أن يقوله الـ model في كل turn، وتشغّل الـ agent، ثم تتحقق بالضبط مما أرسله الـ agent إلى الـ model. بلا network، بلا API keys، بلا flakiness. diff --git a/ar/tool-search.mdx b/ar/tool-search.mdx index 4ce017d..d132170 100644 --- a/ar/tool-search.mdx +++ b/ar/tool-search.mdx @@ -1,7 +1,7 @@ --- title: "Tool search" --- -{/* Generated from docs/tool-search.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/tool-search.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} agent عنده tools كثيرة (عدة سيرفرات MCP أو toolkit داخلية كبيرة) يرسل تعريف كل tool مع كل استدعاء للـ model. هذا يكلّف input tokens، وبعد diff --git a/ar/tools.mdx b/ar/tools.mdx index 158c937..174d7ce 100644 --- a/ar/tools.mdx +++ b/ar/tools.mdx @@ -1,7 +1,7 @@ --- title: "Tools" --- -{/* Generated from docs/tools.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/tools.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـ tool هو دالة typed يستطيع الـ model استدعاءها. عرّفه بـ `defineTool()`: تُستنتج أنواع الوسائط والنتيجة من zod `input`، وتمر الوسائط على validation قبل أن تعمل `execute`، ويمكن وضع النتيجة في أي مكان يقبل tools (`createAgent({ tools: [...] })`، `ToolRegistry.register(tool)`، `AgentBuilder.addTool(tool)`). diff --git a/ar/translations.json b/ar/translations.json index c5e3168..2f2fbd8 100644 --- a/ar/translations.json +++ b/ar/translations.json @@ -1,81 +1,81 @@ { - "acp": "9f443f7f4f8f", - "agent-directories": "e8925c2ace35", - "agent-forge": "54d3c2b742a3", - "agent-spec": "10583c615347", - "ai-sdk-ui": "05fba46ba7ab", - "api-overview": "11eb40db4ecc", - "approvals": "4bc004f33d66", - "auth": "3e0019bfbccc", - "build-a-coding-agent": "fbfa6d9195bf", - "channels": "6e48db30e8a6", - "claude-projects": "47c5a207af7e", - "cli": "c680dac8f533", - "cloudflare-workers": "2da262898537", - "code-mode": "3b97ba05200c", - "coding-agents": "e9e0e98e6dfd", - "compaction": "d87af1f5ba11", - "comparison": "088891111b2f", - "configuration": "9c3d136710cc", - "connectors": "d1974d8c7cfb", - "decisions": "ab1bc01e2ca3", - "deployment": "9b14f365e3cb", - "durable-execution": "7c4b4309d9bb", - "errors": "e0f2402e40c5", - "evals": "91db5a3d7f15", - "examples": "f8847abada67", - "examples/coding-agent-workflows": "63917d151ead", - "examples/connectors": "2b68c9860468", - "examples/heal-benchmark": "f4a5dbf3c7b1", - "examples/hostinger-deploy": "f9da62e3a364", - "examples/phase-pipeline": "702b9bf538c5", - "examples/waves": "c298e2ae94b6", - "executor-api": "955d1a4f4f44", - "flows": "334860c14f68", - "guardrails": "0530b0a01591", - "handoffs": "bda98b7ba264", - "hooks": "d54027280b46", - "hosted-tools": "14eb5f7c8eb9", - "hostinger-vps": "bfa034b8c3be", - "improve": "fe7ce62c0c90", - "installation": "6f6ebfa0d752", + "acp": "772532680313", + "agent-directories": "c02b43f4edc8", + "agent-forge": "f3f7de26b5ce", + "agent-spec": "ce14a7051eb0", + "ai-sdk-ui": "15d6b27d60ed", + "api-overview": "1d0374c2a05c", + "approvals": "aecd1aedcd51", + "auth": "96d49d2fbc57", + "build-a-coding-agent": "66a1c9747054", + "channels": "3c34a1e9e6dc", + "claude-projects": "9ef693862aa7", + "cli": "4d68dd962512", + "cloudflare-workers": "2eb8d3c0be32", + "code-mode": "58bd5f80d95c", + "coding-agents": "2eb5d3381f17", + "compaction": "f586518bf4d1", + "comparison": "ced2205ed15a", + "configuration": "b5c7c00a66b1", + "connectors": "2a05907d0ceb", + "decisions": "1358426f966c", + "deployment": "754e082abe47", + "durable-execution": "47fe70adf0ac", + "errors": "58fd3501283d", + "evals": "e2fb2a318840", + "examples": "435cc31c4968", + "examples/coding-agent-workflows": "7d3ab35f838d", + "examples/connectors": "707be37a1cd8", + "examples/heal-benchmark": "4e73915d4b8b", + "examples/hostinger-deploy": "28eac1585068", + "examples/phase-pipeline": "81370536c732", + "examples/waves": "88c130c67b38", + "executor-api": "31bf85af7df8", + "flows": "d17c4f46364f", + "guardrails": "8dd61738c80c", + "handoffs": "3abb072aef29", + "hooks": "844e57c32f3c", + "hosted-tools": "bcc5cee6e5b4", + "hostinger-vps": "1cec014fd71a", + "improve": "82282d260149", + "installation": "0b34ee7d70da", "introduction": "6d1dd271727f", - "mcp-integration": "bdf06a9a7df2", - "memory": "736c6c0ab8d7", - "migrating-from-ai-sdk": "50a4fd02f6c1", - "migrating-to-create-agent": "d81141c2f492", - "models-and-cost": "37a93593c694", - "nextjs": "0e1085991ea6", - "oauth": "703519381f76", - "observability": "388351398c53", - "openapi-tools": "bd0828b0cdd8", - "pause-and-resume": "8b839f3a0c45", - "permission-modes": "b1ad1d01b071", - "prompting-techniques": "ede9163d12d1", - "providers": "b9a8a9fba50c", - "queue-and-steer": "5c6d0fe80107", - "quickstart": "5e5ce4354c62", - "react": "8c02c305a8f4", - "reasoning": "f493cbaf0d96", - "registry": "d9458f2bb6dc", - "runs": "b83155cb8b2e", - "schedules": "d8fe79433275", - "self-healing": "601ae37e8e12", - "self-healing-paper": "7b6ae560d20c", - "sessions": "38f5c7892813", - "skills": "592829bf0e26", - "stream-events": "10000c736a5d", - "streaming": "7753da6d3911", - "structured-output": "b23a0c90351c", - "sub-agents": "89d71907109a", - "svelte": "402ee543d645", - "testing": "3b9f86347a1f", - "tool-search": "4247e3f941b9", - "tools": "676fa1d59bbe", - "triggers": "8c7109ed08a5", - "troubleshooting": "665092361dfc", - "upgrading": "4f41c0ba06d9", - "utilities": "38d975cefacf", - "vue": "f1ef973fa5b9", - "workspace-tools": "139be9c43d26" + "mcp-integration": "ceddd35a3779", + "memory": "a78bdd50d485", + "migrating-from-ai-sdk": "ed1d8c743a8a", + "migrating-to-create-agent": "cbf8ca40eee8", + "models-and-cost": "fa4b6714a316", + "nextjs": "9ee2b3299217", + "oauth": "ec7d59fb855e", + "observability": "0098d821c467", + "openapi-tools": "2363bc34fe37", + "pause-and-resume": "dac2b170478b", + "permission-modes": "5f7b9bd47122", + "prompting-techniques": "8c8a31943ac6", + "providers": "5b1740c00237", + "queue-and-steer": "91781420753a", + "quickstart": "5337b733f500", + "react": "9386dd5cd96a", + "reasoning": "edc6121173c0", + "registry": "6d63d55c1d5b", + "runs": "2fbfbcce85e9", + "schedules": "e7f15acd33ab", + "self-healing": "c067e595b6b6", + "self-healing-paper": "f991429f43aa", + "sessions": "f78ba8a41465", + "skills": "e14afb010954", + "stream-events": "4088703765bd", + "streaming": "fdeae05d48d7", + "structured-output": "8b4116ce7d5b", + "sub-agents": "b736a6329080", + "svelte": "991936d0fe68", + "testing": "92228e02f375", + "tool-search": "45a88e1af354", + "tools": "cf331871bfd9", + "triggers": "71cfd2d85b69", + "troubleshooting": "78daf1ef3a6f", + "upgrading": "b25b5df1415c", + "utilities": "abc3be75d91f", + "vue": "e63dcea6dd8f", + "workspace-tools": "428231ee4393" } diff --git a/ar/triggers.mdx b/ar/triggers.mdx index 6d10b1c..7b3edb8 100644 --- a/ar/triggers.mdx +++ b/ar/triggers.mdx @@ -1,7 +1,7 @@ --- title: "Triggers" --- -{/* Generated from docs/triggers.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/triggers.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الـtrigger adapter يحوّل آلية تنبيه واحدة إلى run للـagent. الـsubpath `lousho/triggers` هو نقطة التوسعة لآليتك الخاصة (`TriggerAdapter` و`TriggerRegistry`)، ومعه بعض الأدوات المساعدة المشتركة مع الـchannels والـschedules: أشكال `WebhookAuth` و`verifySlackSignature()` و`parseCronExpression()`. أما الـadapters الجاهزة للـwebhook وSlack وcron فقد حُذفت؛ راجع [الانتقال من triggers](/ar/upgrading#الانتقال-من-الـ-triggers). diff --git a/ar/troubleshooting.mdx b/ar/troubleshooting.mdx index 45fff3c..efe6769 100644 --- a/ar/troubleshooting.mdx +++ b/ar/troubleshooting.mdx @@ -1,7 +1,7 @@ --- title: "حل المشاكل" --- -{/* Generated from docs/troubleshooting.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/troubleshooting.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} الصفحة هذه تبدأ من الشيء الذي تراه، وتدلّك على السبب والحل والصفحة التي تشرحه. شغّل أولًا `npx lousho doctor` ([ماذا يفحص](/ar/installation#استكشاف-الأخطاء-lousho-doctor))، وبعدها إذا عندك رمز خطأ (`LOUSHO_...`) ابحث عنه في [الأخطاء](/ar/errors). @@ -35,7 +35,7 @@ title: "حل المشاكل" - **السبب:** الـ quickstart يستخدم `await` على المستوى الأعلى، فالملف يجب أن يكون ES module. - **الحل:** سمِّ الملف `.mts`، أو اضبط `"type": "module"` في `package.json`. -- **للمزيد:** [الـ quickstart في README](https://github.com/LinuxDevil/agent-sdk/blob/main/README.md#quickstart). +- **للمزيد:** [الـ quickstart في README](https://github.com/LinuxDevil/lousho/blob/main/README.md#quickstart). ### إصدار Node قديم جدًا @@ -137,7 +137,7 @@ title: "حل المشاكل" ## طلب مساعدة -افتح issue على [https://github.com/LinuxDevil/agent-sdk/issues](https://github.com/LinuxDevil/agent-sdk/issues). أرفق: +افتح issue على [https://github.com/LinuxDevil/lousho/issues](https://github.com/LinuxDevil/lousho/issues). أرفق: - ناتج `npx lousho doctor --json`، - رمز الخطأ (`LOUSHO_...`) والرسالة كاملة، diff --git a/ar/upgrading.mdx b/ar/upgrading.mdx index 2f7a736..32a394b 100644 --- a/ar/upgrading.mdx +++ b/ar/upgrading.mdx @@ -2,7 +2,7 @@ title: "الترقية إلى 1.0" sidebarTitle: "الترقية إلى 1.0" --- -{/* Generated from docs/upgrading.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/upgrading.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## من يحتاج هذه الصفحة diff --git a/ar/utilities.mdx b/ar/utilities.mdx index e675218..4871906 100644 --- a/ar/utilities.mdx +++ b/ar/utilities.mdx @@ -2,7 +2,7 @@ title: "أدوات مساعدة" description: "دوال مساعدة يصدّرها المسار الفرعي lousho/utils: تشفير و hashing، وتخزين ملفات للمرفقات. لا تحتاج إلى أي منها لبناء agent؛ فهي موجودة للتطبيقات المحيطة به." --- -{/* Generated from docs/utilities.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/utilities.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## التشفير و hashing diff --git a/ar/vue.mdx b/ar/vue.mdx index 0e4aa28..a3f6d45 100644 --- a/ar/vue.mdx +++ b/ar/vue.mdx @@ -1,7 +1,7 @@ --- title: "Vue" --- -{/* Generated from docs/vue.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/vue.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `useLoushoAgent()` في Vue 3 هو [الـ hook الخاص بـReact](/ar/react) نفسه، لكن على شكل composable: المصادر نفسها، والحالة نفسها، والأوامر نفسها، فوق الـ diff --git a/ar/workspace-tools.mdx b/ar/workspace-tools.mdx index 66215f6..6e053fb 100644 --- a/ar/workspace-tools.mdx +++ b/ar/workspace-tools.mdx @@ -1,7 +1,7 @@ --- title: "Workspace tools" --- -{/* Generated from docs/workspace-tools.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/workspace-tools.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} تعطي الـworkspace tools الـagent نظام ملفات وshell، فيستطيع العمل كـcoding agent بأسلوب Claude Code. الأدوات نفسها لا تستخدم `node:fs` ولا `node:child_process`، بل تستدعي واجهتين صغيرتين: `FsProvider` و`ShellProvider`. لذلك تعمل الأدوات نفسها على مجلد محلي، أو على شجرة ملفات في الذاكرة أثناء الاختبارات، أو على container في Docker، أو على sandbox بعيد مثل E2B أو Daytona أو Cloudflare. لشرح عملي كامل، راجع [ابنِ coding agent](/ar/build-a-coding-agent). diff --git a/auth.mdx b/auth.mdx index 506f149..b9e3d32 100644 --- a/auth.mdx +++ b/auth.mdx @@ -1,7 +1,7 @@ --- title: "Route auth and principals" --- -{/* Generated from docs/auth.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/auth.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} An agent served over HTTP needs to know two things about each request: may it call at all, and who is it. `lousho/auth` answers both. You diff --git a/build-a-coding-agent.mdx b/build-a-coding-agent.mdx index 891ca2b..47503b7 100644 --- a/build-a-coding-agent.mdx +++ b/build-a-coding-agent.mdx @@ -2,7 +2,7 @@ title: "Build a coding agent" sidebarTitle: "Build a coding agent" --- -{/* Generated from docs/build-a-coding-agent.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/build-a-coding-agent.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} This page builds a terminal coding agent in the style of Claude Code: it reads and edits files in your project, runs commands, asks you before it writes a diff --git a/channels.mdx b/channels.mdx index d7db485..fa3a26e 100644 --- a/channels.mdx +++ b/channels.mdx @@ -1,7 +1,7 @@ --- title: "Channels" --- -{/* Generated from docs/channels.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/channels.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A **channel** connects an agent to one surface: a JSON API, a webhook, a chat app. It says how an inbound request is authenticated, which conversation it diff --git a/claude-projects.mdx b/claude-projects.mdx index de829c5..a2eaad5 100644 --- a/claude-projects.mdx +++ b/claude-projects.mdx @@ -1,7 +1,7 @@ --- title: "Claude Code projects" --- -{/* Generated from docs/claude-projects.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/claude-projects.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A repository that already carries Claude Code's conventional files — `CLAUDE.md`, `.claude/skills/`, `.claude/agents/`, `.claude/rules/`, diff --git a/cli.mdx b/cli.mdx index 1eb44d6..d46bba0 100644 --- a/cli.mdx +++ b/cli.mdx @@ -1,7 +1,7 @@ --- title: "CLI" --- -{/* Generated from docs/cli.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/cli.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Installing the package also installs the `lousho` command. It scaffolds projects, checks your setup, runs an agent locally, serves it over MCP, runs diff --git a/cloudflare-workers.mdx b/cloudflare-workers.mdx index 9bf80a9..00827d0 100644 --- a/cloudflare-workers.mdx +++ b/cloudflare-workers.mdx @@ -1,7 +1,7 @@ --- title: "Cloudflare Workers" --- -{/* Generated from docs/cloudflare-workers.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/cloudflare-workers.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `npx lousho build --target=cloudflare-worker --agent=agent.yaml` generates a module Worker (`worker.ts`), a `wrangler.toml` and a bundle, `dist/worker.js`, that the build checks for `node:` imports. It needs `tsup` and `wrangler` installed, and it takes an [agent spec file](/configuration) or an [agent directory](/agent-directories) (`npx lousho build ./my-agent --target=cloudflare-worker`). The Worker serves the same [HTTP API](/deployment#http-api) as the other targets. This target has the tightest limits of the three, so they come first. diff --git a/code-mode.mdx b/code-mode.mdx index 0ff2aa6..6ac85bc 100644 --- a/code-mode.mdx +++ b/code-mode.mdx @@ -1,7 +1,7 @@ --- title: "Code mode" --- -{/* Generated from docs/code-mode.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/code-mode.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} With code mode the model gets one more tool, `run_code`. Instead of calling tools one at a time, with a model round trip after each, it writes a short diff --git a/coding-agents.mdx b/coding-agents.mdx index 6599103..6bf362f 100644 --- a/coding-agents.mdx +++ b/coding-agents.mdx @@ -10,7 +10,7 @@ A coding agent writes better Lousho code when it reads the current docs instead | File | Where | What it is | | ---- | ----- | ---------- | | `llms.txt` | In the npm package: `node_modules/lousho/llms.txt` | An [llmstxt.org](https://llmstxt.org) index: one line per docs page, with a summary and a link | -| `llms-full.txt` | In the [SDK repository](https://github.com/LinuxDevil/agent-sdk/blob/main/llms-full.txt), not in the npm package (it is about 1 MB) | The README and every docs page in one file, with absolute links | +| `llms-full.txt` | In the [SDK repository](https://github.com/LinuxDevil/lousho/blob/main/llms-full.txt), not in the npm package (it is about 1 MB) | The README and every docs page in one file, with absolute links | Both are generated from the same docs as this site and checked in the SDK's CI. The `llms.txt` in `node_modules` matches the version you have installed; the repository's `llms-full.txt` follows the SDK's `main` branch, which can be ahead of the release you installed. Tell your coding agent where they are, for example in your project's `AGENTS.md` or `CLAUDE.md`: diff --git a/compaction.mdx b/compaction.mdx index 0e056f7..9c18fb5 100644 --- a/compaction.mdx +++ b/compaction.mdx @@ -1,7 +1,7 @@ --- title: "Context compaction" --- -{/* Generated from docs/compaction.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/compaction.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A long agent run fills its context window mostly with tool results: file contents, search hits, API responses the model read once and no longer needs diff --git a/comparison.mdx b/comparison.mdx index bcf1518..5c23ede 100644 --- a/comparison.mdx +++ b/comparison.mdx @@ -2,10 +2,10 @@ title: "How Lousho compares" description: "This page compares Lousho with four other TypeScript agent libraries: the Vercel AI SDK (ToolLoopAgent), the OpenAI Agents SDK for JavaScript, Mastra and LangGraph.js. Each competitor cell comes from that project's official docs." --- -{/* Generated from docs/comparison.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/comparison.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Last checked: 9 Nov 2026. These projects move fast, so a cell may be out of -date. Corrections welcome: [open an issue](https://github.com/LinuxDevil/agent-sdk/issues). +date. Corrections welcome: [open an issue](https://github.com/LinuxDevil/lousho/issues). ## How to read the table diff --git a/configuration.mdx b/configuration.mdx index eb1cba6..2e84194 100644 --- a/configuration.mdx +++ b/configuration.mdx @@ -1,7 +1,7 @@ --- title: "Configuration" --- -{/* Generated from docs/configuration.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/configuration.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## Agent spec files (`AgentSpec`) diff --git a/connectors.mdx b/connectors.mdx index 3ba2ed9..4e9aa3d 100644 --- a/connectors.mdx +++ b/connectors.mdx @@ -1,7 +1,7 @@ --- title: "Connectors" --- -{/* Generated from docs/connectors.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/connectors.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Connectors are how the agent reaches external services — Google Docs, Slack, GitHub, Notion, a database — the way Claude's connectors do. In Lousho a diff --git a/decisions.mdx b/decisions.mdx index a7a2874..bebc6cc 100644 --- a/decisions.mdx +++ b/decisions.mdx @@ -1,7 +1,7 @@ --- title: "Decisions" --- -{/* Generated from docs/decisions.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/decisions.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `decide()` calls the OpenAI Decisions API (`POST /v1/decisions`): a purpose-built endpoint that evaluates shared input against typed questions and returns calibrated answers roughly 10x faster than a full model turn. It is the fast, deterministic shape for the typed-decision pattern — classify, route, gate, score — without spinning up an agent run. @@ -114,4 +114,4 @@ Failures are `SDKError`s: missing key is `LOUSHO_PROVIDER_MISSING_API_KEY`, a `4 - **[Structured output](/structured-output)** (`createAgent({ output })`) — when you need your own JSON schema: extracted fields, a written explanation, or an object shaped like your domain. Works on any provider, including strict structured-output endpoints (see [Schema limits](/structured-output#schema-limits)). - **[Sub-agents](/sub-agents)** or [flows](/flows) — when the decision needs tools or multi-step reasoning before it answers. -The [coding-agent-workflows example](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/coding-agent-workflows) shows the same route/confidence/floor pattern built on `output` schemas — swap its triage step for `decide()` when you are on OpenAI and want the dedicated endpoint. [Prompting techniques](/prompting-techniques#typed-decisions-system-one-style) places both forms in the wider typed-decision catalog. +The [coding-agent-workflows example](https://github.com/LinuxDevil/lousho/tree/main/examples/coding-agent-workflows) shows the same route/confidence/floor pattern built on `output` schemas — swap its triage step for `decide()` when you are on OpenAI and want the dedicated endpoint. [Prompting techniques](/prompting-techniques#typed-decisions-system-one-style) places both forms in the wider typed-decision catalog. diff --git a/deployment.mdx b/deployment.mdx index e1d698a..99c6b67 100644 --- a/deployment.mdx +++ b/deployment.mdx @@ -1,7 +1,7 @@ --- title: "Deployment" --- -{/* Generated from docs/deployment.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/deployment.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `lousho build` turns an agent spec file (see [Configuration](/configuration)) into a deployable artifact for one target platform: diff --git a/docs.json b/docs.json index 729c8fb..7fdba68 100644 --- a/docs.json +++ b/docs.json @@ -29,7 +29,7 @@ ], "primary": { "type": "github", - "href": "https://github.com/LinuxDevil/agent-sdk" + "href": "https://github.com/LinuxDevil/lousho" } }, "navigation": { @@ -411,7 +411,7 @@ }, "footer": { "socials": { - "github": "https://github.com/LinuxDevil/agent-sdk" + "github": "https://github.com/LinuxDevil/lousho" } } } diff --git a/durable-execution.mdx b/durable-execution.mdx index c665fa4..6eaa83d 100644 --- a/durable-execution.mdx +++ b/durable-execution.mdx @@ -1,7 +1,7 @@ --- title: "Durable execution" --- -{/* Generated from docs/durable-execution.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/durable-execution.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Give `createAgent()` a `store` and pass a `sessionId` to `send()`, and the run survives a crash, a restart, an abort or a human approval that takes diff --git a/errors.mdx b/errors.mdx index f603f40..21c3eb0 100644 --- a/errors.mdx +++ b/errors.mdx @@ -2,7 +2,7 @@ title: "Errors" description: "Every deliberate SDK error is an SDKError with a stable LOUSHO_* code, a hint and a link to its section here; this page lists each code with its cause and fix." --- -{/* Generated from docs/errors.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/errors.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Nearly every error the SDK throws on purpose is an `SDKError` (or a subclass such as `ConfigurationError`, `ValidationError` or `MissingPeerDependencyError`) diff --git a/evals.mdx b/evals.mdx index 55d6d3e..97d55b9 100644 --- a/evals.mdx +++ b/evals.mdx @@ -1,7 +1,7 @@ --- title: "Evals" --- -{/* Generated from docs/evals.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/evals.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Evals are regression tests for agent *behaviour*: which tools the agent called, in what order, with which arguments, how many steps it took, and what it said. diff --git a/examples.mdx b/examples.mdx index 76d14ab..48bc9c4 100644 --- a/examples.mdx +++ b/examples.mdx @@ -3,17 +3,17 @@ title: "Examples" description: "Runnable examples from the SDK repository. Most use the built-in mock provider, so they need no API key." --- -The examples live in the [`examples/`](https://github.com/LinuxDevil/agent-sdk/tree/main/examples) directory of the SDK repository. Clone it, install, and run any of them with the command shown: +The examples live in the [`examples/`](https://github.com/LinuxDevil/lousho/tree/main/examples) directory of the SDK repository. Clone it, install, and run any of them with the command shown: ```bash -git clone https://github.com/LinuxDevil/agent-sdk.git +git clone https://github.com/LinuxDevil/lousho.git cd agent-sdk npm install ``` ## Flagship - + A monitoring alert reaches a monitor agent, which delegates a fix behind a human approval gate. A Slack "Fix it" button resumes the paused run, a fixer agent writes a patch, and guardrails check the patch before a GitHub pull request is opened. Runs against mocks by default: `npm run pipeline:demo`. @@ -21,25 +21,25 @@ npm install | Example | What it shows | Run it | | ------- | ------------- | ------ | -| [coding-harness](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/coding-harness) | A production-shaped coding agent: workspace tools with checkpoints, permission rules, a loop guard, a cost cap, and the `coding-kit` / `coding-pi` registry kits it ships as | `npx tsx examples/coding-harness/index.ts` | -| [deep-research](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/deep-research) | The multi-agent research archetype: a coordinator fans out parallel `task` calls to researcher sub-agents, then synthesizes a cited report | `npm run example:deep-research` | -| [support-desk](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/support-desk) | A triage agent that `handoff()`s the conversation to billing/tech specialists, with an approval-gated refund tool | `npm run example:support-desk` | -| [evaluator-loop](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/evaluator-loop) | The writer/critic archetype: a draft is scored by `llmCritique()` and revised against its feedback until it passes | `npm run example:evaluator-loop` | -| [data-analyst](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/data-analyst) | Text-to-SQL over `node:sqlite` with defense-in-depth read-only enforcement — `DELETE`/`DROP` can never run | `npm run example:data-analyst` | -| [plan-mode](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/plan-mode) | A coding agent that reads and proposes in `plan` mode, then applies edits after `session.setPermissionMode('acceptEdits')` | `npm run example:plan-mode` | +| [coding-harness](https://github.com/LinuxDevil/lousho/tree/main/examples/coding-harness) | A production-shaped coding agent: workspace tools with checkpoints, permission rules, a loop guard, a cost cap, and the `coding-kit` / `coding-pi` registry kits it ships as | `npx tsx examples/coding-harness/index.ts` | +| [deep-research](https://github.com/LinuxDevil/lousho/tree/main/examples/deep-research) | The multi-agent research archetype: a coordinator fans out parallel `task` calls to researcher sub-agents, then synthesizes a cited report | `npm run example:deep-research` | +| [support-desk](https://github.com/LinuxDevil/lousho/tree/main/examples/support-desk) | A triage agent that `handoff()`s the conversation to billing/tech specialists, with an approval-gated refund tool | `npm run example:support-desk` | +| [evaluator-loop](https://github.com/LinuxDevil/lousho/tree/main/examples/evaluator-loop) | The writer/critic archetype: a draft is scored by `llmCritique()` and revised against its feedback until it passes | `npm run example:evaluator-loop` | +| [data-analyst](https://github.com/LinuxDevil/lousho/tree/main/examples/data-analyst) | Text-to-SQL over `node:sqlite` with defense-in-depth read-only enforcement — `DELETE`/`DROP` can never run | `npm run example:data-analyst` | +| [plan-mode](https://github.com/LinuxDevil/lousho/tree/main/examples/plan-mode) | A coding agent that reads and proposes in `plan` mode, then applies edits after `session.setPermissionMode('acceptEdits')` | `npm run example:plan-mode` | | [waves](/examples/waves) | Waves: bounded parallel workers, a deterministic aggregate, a typed verify gate, and extension waves seeded by the gaps it names | `npm run example:waves` | | [phase-pipeline](/examples/phase-pipeline) | Phase engineering on flows: discuss → spec → plan → implement → proof gate, with the retry ladder as a `oneOf` chain | `npm run example:phase-pipeline` | | [coding-agent-workflows](/examples/coding-agent-workflows) | A coding agent routed by a typed decision (`{ route, confidence }` via `output`), with a confidence floor escalating to the careful path | `npm run example:coding-agent-workflows` | | [connectors](/examples/connectors) | MCP connectors (the Claude-connectors shape) run fully offline against a mock `serveMcp` stdio server | `npm run example:connectors` | | [hostinger-deploy](/examples/hostinger-deploy) | Deploy an agent to a real Hostinger VPS: API verification, `docker save` over SSH, health + approval smoke tests | `node examples/hostinger-deploy/deploy.mjs --vm ` | -| [agent-dir](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/agent-dir) | An agent defined as a directory (`instructions.md`, `tools/`, `skills/`, `agent.json`) and loaded with `loadAgentDir()` | `npm run example:agent-dir` | -| [support-bot](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/support-bot) | A minimal customer-support agent | `npm run example:support-bot` | -| [research-assistant](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/research-assistant) | A research agent with the built-in `http` tool | `npm run example:research-assistant` | -| [doc-qa](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/doc-qa) | Question answering scoped to one document | `npm run example:doc-qa` | -| [workflow-router](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/workflow-router) | Classifying a request into a fixed set of categories | `npm run example:workflow-router` | -| [slack-notifier](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/slack-notifier) | Turning a raw event description into a short Slack notification | `npm run example:slack-notifier` | -| [tracing](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/tracing) | A run wired to console and OpenTelemetry trace exporters | `npm run example:tracing:console` | -| [openrouter](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/openrouter) | `OpenRouterProvider`: generation, streaming, tool calling, model listing. Needs `OPENROUTER_API_KEY` | `npm run example:openrouter` | +| [agent-dir](https://github.com/LinuxDevil/lousho/tree/main/examples/agent-dir) | An agent defined as a directory (`instructions.md`, `tools/`, `skills/`, `agent.json`) and loaded with `loadAgentDir()` | `npm run example:agent-dir` | +| [support-bot](https://github.com/LinuxDevil/lousho/tree/main/examples/support-bot) | A minimal customer-support agent | `npm run example:support-bot` | +| [research-assistant](https://github.com/LinuxDevil/lousho/tree/main/examples/research-assistant) | A research agent with the built-in `http` tool | `npm run example:research-assistant` | +| [doc-qa](https://github.com/LinuxDevil/lousho/tree/main/examples/doc-qa) | Question answering scoped to one document | `npm run example:doc-qa` | +| [workflow-router](https://github.com/LinuxDevil/lousho/tree/main/examples/workflow-router) | Classifying a request into a fixed set of categories | `npm run example:workflow-router` | +| [slack-notifier](https://github.com/LinuxDevil/lousho/tree/main/examples/slack-notifier) | Turning a raw event description into a short Slack notification | `npm run example:slack-notifier` | +| [tracing](https://github.com/LinuxDevil/lousho/tree/main/examples/tracing) | A run wired to console and OpenTelemetry trace exporters | `npm run example:tracing:console` | +| [openrouter](https://github.com/LinuxDevil/lousho/tree/main/examples/openrouter) | `OpenRouterProvider`: generation, streaming, tool calling, model listing. Needs `OPENROUTER_API_KEY` | `npm run example:openrouter` | ## Related guides diff --git a/examples/coding-agent-workflows.mdx b/examples/coding-agent-workflows.mdx index a52e984..e7b4e95 100644 --- a/examples/coding-agent-workflows.mdx +++ b/examples/coding-agent-workflows.mdx @@ -3,7 +3,7 @@ title: "coding-agent-workflows" sidebarTitle: "Coding agent workflows" description: "A coding agent driven by deterministic workflows whose routing is a typed decision (the \"System One\" pattern)." --- -{/* Generated from examples/coding-agent-workflows/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/coding-agent-workflows/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A typed decision answers a structured probabilistic question — classify, route, score, verify — instead of generating text. The Lousho form of that is `createAgent({ output: DECISION })`: the model returns `{ route, confidence, reason }` validated by a zod schema (readable as `result.object`), and the workflow branches on the typed fields. diff --git a/examples/connectors.mdx b/examples/connectors.mdx index 7d1c59c..f697b87 100644 --- a/examples/connectors.mdx +++ b/examples/connectors.mdx @@ -3,7 +3,7 @@ title: "connectors" sidebarTitle: "Connectors" description: "Connect an agent to external services (Google Docs, Slack, GitHub, Notion, …) the way Claude's connectors do: over MCP." --- -{/* Generated from examples/connectors/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/connectors/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ```ts const agent = createAgent({ diff --git a/examples/heal-benchmark.mdx b/examples/heal-benchmark.mdx index a43a4cf..c55de43 100644 --- a/examples/heal-benchmark.mdx +++ b/examples/heal-benchmark.mdx @@ -3,7 +3,7 @@ title: "heal-benchmark" sidebarTitle: "Heal benchmark" description: "A repeatable way to tell whether a change to an agent's harness helped. It runs four agents through 28 tasks and reports each harness version's pass rate, noise, tokens and tool calls." --- -{/* Generated from examples/heal-benchmark/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/heal-benchmark/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} | Agent | What it tests | | --- | --- | diff --git a/examples/hostinger-deploy.mdx b/examples/hostinger-deploy.mdx index ec893e1..3b49bae 100644 --- a/examples/hostinger-deploy.mdx +++ b/examples/hostinger-deploy.mdx @@ -2,7 +2,7 @@ title: "hostinger-deploy" sidebarTitle: "Hostinger deploy" --- -{/* Generated from examples/hostinger-deploy/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/hostinger-deploy/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Deploy a Lousho agent to a real VPS: `agent/` is an agent directory for a small ops assistant (`vps-ops`), and `deploy.mjs` builds it with @@ -52,4 +52,4 @@ curl -N http://:3000/chat \ Port 3000 is plain HTTP — put Traefik (installed on the VPS template) or another TLS terminator in front for anything real, per -[the deployment docs](https://github.com/LinuxDevil/agent-sdk/blob/main/docs/deployment.md). +[the deployment docs](https://github.com/LinuxDevil/lousho/blob/main/docs/deployment.md). diff --git a/examples/phase-pipeline.mdx b/examples/phase-pipeline.mdx index ce172a6..342ec56 100644 --- a/examples/phase-pipeline.mdx +++ b/examples/phase-pipeline.mdx @@ -2,7 +2,7 @@ title: "phase-pipeline" sidebarTitle: "Phase pipeline" --- -{/* Generated from examples/phase-pipeline/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/phase-pipeline/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} **Phase engineering** — a deterministic phase pipeline where the model fills each phase but the *order*, the *gate*, and what counts as *done* belong to the harness. The shape behind agentic-workflow repos (phases-oss, leeovery/agentic-workflows): diff --git a/examples/waves.mdx b/examples/waves.mdx index 428c2d1..8cd1b0b 100644 --- a/examples/waves.mdx +++ b/examples/waves.mdx @@ -2,7 +2,7 @@ title: "waves" sidebarTitle: "Waves" --- -{/* Generated from examples/waves/README.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from examples/waves/README.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} **Waves** — Workers · Aggregate · Verify · Extend — the bounded-parallel pattern from the community "waves" workflows (rayfernando-skills, Wave-Engineering/claudecode-workflow). Unlike a plain task fan-out, the wave owns a *verification spine*: worker outputs are merged deterministically, a verifier judges the aggregate against a typed schema, and a follow-up wave runs **only** on the gaps the verifier names. @@ -36,6 +36,6 @@ npm run example:waves # offline, scripted workers + verifier OPENROUTER_API_KEY=… npx tsx examples/waves/index.ts # live ``` -When to reach for it: breadth-first work where *coverage* is checkable — summarizing a corpus, auditing a codebase, migrating files — and a partial answer is worse than a slow one. When a single orchestrator can hold the whole task, [deep-research](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/deep-research/)'s lead-delegates-subtasks shape is simpler. +When to reach for it: breadth-first work where *coverage* is checkable — summarizing a corpus, auditing a codebase, migrating files — and a partial answer is worse than a slow one. When a single orchestrator can hold the whole task, [deep-research](https://github.com/LinuxDevil/lousho/blob/main/examples/deep-research/)'s lead-delegates-subtasks shape is simpler. See [docs/prompting-techniques.md](/prompting-techniques) for where wave engineering sits in the full pattern catalog. diff --git a/executor-api.mdx b/executor-api.mdx index 0d80c53..a4c3ebe 100644 --- a/executor-api.mdx +++ b/executor-api.mdx @@ -2,7 +2,7 @@ title: "The executor API" sidebarTitle: "Executor API" --- -{/* Generated from docs/executor-api.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/executor-api.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `createAgent()` is the recommended entry point (see the [Quick Start](/quickstart)). This page is for code that needs the lower-level options it does not take. To move existing executor code over to `createAgent()`, see [Migrating to createAgent()](/migrating-to-create-agent). diff --git a/flows.mdx b/flows.mdx index 7ecb8cc..064ee5c 100644 --- a/flows.mdx +++ b/flows.mdx @@ -1,7 +1,7 @@ --- title: "Flows" --- -{/* Generated from docs/flows.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/flows.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A flow is a fixed, multi-step workflow run by one agent: the steps and their order are written down up front instead of being left to the model's judgment. diff --git a/guardrails.mdx b/guardrails.mdx index b442842..5e9d876 100644 --- a/guardrails.mdx +++ b/guardrails.mdx @@ -1,7 +1,7 @@ --- title: "Guardrails and sandboxing" --- -{/* Generated from docs/guardrails.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/guardrails.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Tools for running agents safely: **input and output guardrails** check what goes into and comes out of any run (and its tool arguments), **patch @@ -180,7 +180,7 @@ checked by any guardrail yet. `runPatchChecks(patch, checks)` runs every check concurrently over a proposed patch and rolls the results up into one verdict. The -[ops-pipeline example](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/ops-pipeline) uses it to gate a fixer +[ops-pipeline example](https://github.com/LinuxDevil/lousho/blob/main/examples/ops-pipeline) uses it to gate a fixer agent's patch before a pull request is opened: ```ts diff --git a/handoffs.mdx b/handoffs.mdx index 5ded99b..ade11f3 100644 --- a/handoffs.mdx +++ b/handoffs.mdx @@ -2,7 +2,7 @@ title: "Handoffs" description: "A handoff passes the whole conversation to another agent. A triage agent reads the user's request and hands it to a specialist; the specialist answers the user directly, and in a session it keeps the conversation in later turns." --- -{/* Generated from docs/handoffs.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/handoffs.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## Handoffs or sub-agents? diff --git a/hooks.mdx b/hooks.mdx index e3eefb2..1e06679 100644 --- a/hooks.mdx +++ b/hooks.mdx @@ -1,7 +1,7 @@ --- title: "Hooks" --- -{/* Generated from docs/hooks.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/hooks.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A hook is an object `{ name, preToolCall?, postToolCall?, preGenerate?, postGenerate? }` that runs inside the agent loop, around every tool call and every model call. Use one to write an audit log, deny a tool, redact or truncate a result, inject context before the model call, or change the request (a header, `temperature`) it sends. diff --git a/hosted-tools.mdx b/hosted-tools.mdx index 3c21091..dfdd8f1 100644 --- a/hosted-tools.mdx +++ b/hosted-tools.mdx @@ -1,7 +1,7 @@ --- title: "Hosted provider tools" --- -{/* Generated from docs/hosted-tools.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/hosted-tools.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Some model providers run tools themselves, inside the model request: OpenAI's web search, code interpreter and file search are the common ones. Put them in diff --git a/hostinger-vps.mdx b/hostinger-vps.mdx index a13184e..eb8139d 100644 --- a/hostinger-vps.mdx +++ b/hostinger-vps.mdx @@ -1,11 +1,11 @@ --- title: "Deploying to a Hostinger VPS" --- -{/* Generated from docs/hostinger-vps.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/hostinger-vps.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} The `docker` [deployment](/deployment) target is a plain Dockerfile + bundled server, so it runs on any Docker host — including a Hostinger VPS. -The [hostinger-deploy](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/hostinger-deploy) +The [hostinger-deploy](https://github.com/LinuxDevil/lousho/tree/main/examples/hostinger-deploy) example ships a complete, tested recipe: `deploy.mjs` verifies the VM through the Hostinger API, builds locally, ships the context over SSH, and runs the container — then proves the deploy with `/health` and one real `/chat` turn. diff --git a/improve.mdx b/improve.mdx index 4f623ea..298eaad 100644 --- a/improve.mdx +++ b/improve.mdx @@ -2,7 +2,7 @@ title: "Improving an agent" description: "improveAgent looks for a better harness offline, tests it against the current one, and returns a proposal for a person to review." --- -{/* Generated from docs/improve.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/improve.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `improveAgent()` looks for a better set of instructions and skills for an agent, using failing cases you give it. It only returns a proposal. Nothing is applied, diff --git a/installation.mdx b/installation.mdx index 21e78a2..43b9872 100644 --- a/installation.mdx +++ b/installation.mdx @@ -1,7 +1,7 @@ --- title: "Installation" --- -{/* Generated from docs/installation.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/installation.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## Requirements @@ -227,7 +227,7 @@ nothing (an npm quirk with absolute drive-letter paths). Use a relative path (`npx ./create-lousho-.tgz my-agent`) or a `file:` spec (`npx file:C:/abs/path/create-lousho-.tgz my-agent`) instead. -`npm install github:LinuxDevil/agent-sdk` does **not** work: `dist/` is not in +`npm install github:LinuxDevil/lousho` does **not** work: `dist/` is not in git and the repository has no `prepare` build step, so the install has no entry points. diff --git a/mcp-integration.mdx b/mcp-integration.mdx index cf3daa2..32a40e6 100644 --- a/mcp-integration.mdx +++ b/mcp-integration.mdx @@ -2,7 +2,7 @@ title: "MCP (Model Context Protocol)" sidebarTitle: "MCP" --- -{/* Generated from docs/mcp.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/mcp.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} MCP works in two directions. Your agent can call the tools of MCP servers, or your agent can be an MCP server that Claude Code, Cursor and other MCP clients diff --git a/memory.mdx b/memory.mdx index 4743e9d..3979d9e 100644 --- a/memory.mdx +++ b/memory.mdx @@ -1,7 +1,7 @@ --- title: "Memory" --- -{/* Generated from docs/memory.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/memory.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A [session](/sessions) remembers one conversation. Memory is what an agent keeps **across** conversations: a user's preferences, facts it was told, diff --git a/migrating-from-ai-sdk.mdx b/migrating-from-ai-sdk.mdx index 5fbd61c..a1e32c5 100644 --- a/migrating-from-ai-sdk.mdx +++ b/migrating-from-ai-sdk.mdx @@ -2,7 +2,7 @@ title: "Migrating from the Vercel AI SDK" sidebarTitle: "Migrating from the AI SDK" --- -{/* Generated from docs/migrating-from-ai-sdk.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/migrating-from-ai-sdk.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} This page is for code built on the Vercel AI SDK's agent loop: `generateText({ tools })` or `streamText({ tools })` with a step limit, or `new ToolLoopAgent({ ... })`. You can keep diff --git a/migrating-to-create-agent.mdx b/migrating-to-create-agent.mdx index f7b72ab..984d868 100644 --- a/migrating-to-create-agent.mdx +++ b/migrating-to-create-agent.mdx @@ -2,7 +2,7 @@ title: "Migrating to createAgent()" sidebarTitle: "Migrating to createAgent()" --- -{/* Generated from docs/migrating-to-create-agent.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/migrating-to-create-agent.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} This page is for code that uses `AgentBuilder`, `AgentExecutor`, `ToolRegistry` and `resumeAfterApproval()`: the lower-level API the SDK started with. Those exports now diff --git a/models-and-cost.mdx b/models-and-cost.mdx index 035d73d..30e2b8f 100644 --- a/models-and-cost.mdx +++ b/models-and-cost.mdx @@ -2,7 +2,7 @@ title: "Models, tokens and cost" sidebarTitle: "Models and cost" --- -{/* Generated from docs/models-and-cost.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/models-and-cost.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## Models and the price table diff --git a/nextjs.mdx b/nextjs.mdx index 47f55e3..a45ef0e 100644 --- a/nextjs.mdx +++ b/nextjs.mdx @@ -2,7 +2,7 @@ title: "Next.js and other Fetch-API frameworks" sidebarTitle: "Next.js and Fetch frameworks" --- -{/* Generated from docs/nextjs.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/nextjs.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `createRouteHandler(agent, options?)` serves an agent from a route file of any framework whose handlers take a Web `Request` and return a `Response`: the diff --git a/oauth.mdx b/oauth.mdx index f24b19d..9fae594 100644 --- a/oauth.mdx +++ b/oauth.mdx @@ -1,7 +1,7 @@ --- title: "OAuth" --- -{/* Generated from docs/oauth.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/oauth.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Tools and MCP servers that call an API on someone's behalf need an OAuth access token, and usually a refresh token to get the next one. This page covers diff --git a/observability.mdx b/observability.mdx index f4ed34d..5d41bc4 100644 --- a/observability.mdx +++ b/observability.mdx @@ -2,7 +2,7 @@ title: "Tracing and observability" sidebarTitle: "Tracing" --- -{/* Generated from docs/observability.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/observability.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `AgentExecutor.execute()` and `FlowExecutor.execute()` emit spans through a `TraceExporter`. Span names and attributes follow the diff --git a/openapi-tools.mdx b/openapi-tools.mdx index 7c9878c..032aa2a 100644 --- a/openapi-tools.mdx +++ b/openapi-tools.mdx @@ -1,7 +1,7 @@ --- title: "OpenAPI tools" --- -{/* Generated from docs/openapi-tools.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/openapi-tools.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Most services publish an OpenAPI document and no MCP server. `openApiTools()` turns that document into one typed tool per API operation, so an agent can call diff --git a/pause-and-resume.mdx b/pause-and-resume.mdx index 93bd80f..3a74da3 100644 --- a/pause-and-resume.mdx +++ b/pause-and-resume.mdx @@ -2,7 +2,7 @@ title: "Pause, restart, resume" description: "A tutorial for one job: an agent that stops to ask a human before a risky tool call, survives a process restart while it waits, and carries on once someone decides. Then what happens when the process dies in the middle of a tool." --- -{/* Generated from docs/pause-and-resume.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/pause-and-resume.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} The snippets use `mockModel` so they run as pasted, with no API key. Swap in a real `provider` or `model` for your own agent; nothing else changes. Each diff --git a/permission-modes.mdx b/permission-modes.mdx index 02ccb72..c04e061 100644 --- a/permission-modes.mdx +++ b/permission-modes.mdx @@ -1,7 +1,7 @@ --- title: "Permission modes" --- -{/* Generated from docs/permission-modes.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/permission-modes.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A permission mode puts an agent in a named state instead of writing rules: look but do not touch (`plan`), edit files without asking (`acceptEdits`), or never diff --git a/prompting-techniques.mdx b/prompting-techniques.mdx index e8fc2df..9f454de 100644 --- a/prompting-techniques.mdx +++ b/prompting-techniques.mdx @@ -2,7 +2,7 @@ title: "Prompting and context techniques" description: "Agent quality comes from three layers, and it is worth knowing which layer a technique lives at, because the fix for a misbehaving agent is different at each one:" --- -{/* Generated from docs/prompting-techniques.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/prompting-techniques.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} 1. **Prompt-level techniques** — what a single model call is asked to do and how the ask is phrased. @@ -26,11 +26,11 @@ These all reduce to instructions and message shape — `createAgent`'s | Few-shot (in-context learning) | Show input/output pairs so the model copies the shape | Put example pairs in `instructions`, or seed `agent.session().send()` turns — a scripted "user/assistant" exchange the real turn continues | | Chain-of-thought | "Think step by step" before answering | Instructions. Prefer `reasoning` effort where the provider supports it — reasoning tokens are cleaner than prompted monologue | | Self-consistency | Sample N answers, take the majority | N runs + a vote step (a flow `parallel`/`llmCall` fan-out, or a sub-agent per sample) | -| Tree/graph-of-thoughts | Explore a branching search of reasoning paths | A sub-agent per branch, an aggregator comparing them — same fan-out shape as [deep-research](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/deep-research) | +| Tree/graph-of-thoughts | Explore a branching search of reasoning paths | A sub-agent per branch, an aggregator comparing them — same fan-out shape as [deep-research](https://github.com/LinuxDevil/lousho/tree/main/examples/deep-research) | | ReAct | Interleave reasoning and acting | The agent loop itself: every Lousho tool loop is ReAct | -| Reflexion | Critique the last attempt, retry with the critique | [evaluator-loop](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/evaluator-loop): `llmCritique` feedback carried into the next draft | +| Reflexion | Critique the last attempt, retry with the critique | [evaluator-loop](https://github.com/LinuxDevil/lousho/tree/main/examples/evaluator-loop): `llmCritique` feedback carried into the next draft | | Least-to-most | Solve the easy sub-problem first, chain to harder ones | Instructions decomposing into ordered steps, or a `flows` sequence | -| Plan-and-solve / plan-and-execute | Write a plan, then execute it | `permissionMode: 'plan'` — the agent must produce a plan before any write tool runs; see [plan-mode](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/plan-mode) | +| Plan-and-solve / plan-and-execute | Write a plan, then execute it | `permissionMode: 'plan'` — the agent must produce a plan before any write tool runs; see [plan-mode](https://github.com/LinuxDevil/lousho/tree/main/examples/plan-mode) | | Skeleton-of-thought | Outline first, expand each point (in parallel) | A `task` fan-out over outline points — the deep-research shape | | Self-ask | The model asks and answers its own follow-ups | Instructions, or an explicit `oneOf` flow that loops a clarifying sub-agent | | Generated knowledge | Ask the model to state relevant facts, then answer | A first `llmCall` into a variable, the second prompt templated with it | @@ -91,14 +91,14 @@ the community terms that named two of them — all runnable in the repo: | Pattern | Shape | Lousho example | | ------- | ----- | -------------- | | Prompt chaining | Step → step → step, each a call | `flows` sequences (`llmCall` → `llmCall`) | -| Routing | Classify, then dispatch to a specialist | [workflow-router](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/workflow-router); durable version: [support-desk](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/support-desk) handoffs | +| Routing | Classify, then dispatch to a specialist | [workflow-router](https://github.com/LinuxDevil/lousho/tree/main/examples/workflow-router); durable version: [support-desk](https://github.com/LinuxDevil/lousho/tree/main/examples/support-desk) handoffs | | Parallelization | N calls at once, results merged | Parallel `task` calls (`toolConcurrency` is unbounded by default) | -| Orchestrator-workers | A lead decomposes work, workers execute | [deep-research](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/deep-research) | -| Evaluator-optimizer | Draft → score → revise loop | [evaluator-loop](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/evaluator-loop) (`llmCritique`) | -| **Wave engineering** | Bounded waves of parallel workers with a verify gate and scoped extension (WAVES: Workers · Aggregate · Verify · Extend) | [waves](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/waves) | -| **Phase engineering** | Deterministic phase pipelines — spec → plan → implement → verify — the model can't skip, with proof-command gates | [phase-pipeline](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/phase-pipeline) (flows + real proof steps) | +| Orchestrator-workers | A lead decomposes work, workers execute | [deep-research](https://github.com/LinuxDevil/lousho/tree/main/examples/deep-research) | +| Evaluator-optimizer | Draft → score → revise loop | [evaluator-loop](https://github.com/LinuxDevil/lousho/tree/main/examples/evaluator-loop) (`llmCritique`) | +| **Wave engineering** | Bounded waves of parallel workers with a verify gate and scoped extension (WAVES: Workers · Aggregate · Verify · Extend) | [waves](https://github.com/LinuxDevil/lousho/tree/main/examples/waves) | +| **Phase engineering** | Deterministic phase pipelines — spec → plan → implement → verify — the model can't skip, with proof-command gates | [phase-pipeline](https://github.com/LinuxDevil/lousho/tree/main/examples/phase-pipeline) (flows + real proof steps) | | Autonomous loop | Open-ended tool loop until done | The default agent loop, bounded by `maxSteps`, `limits`, permissions | -| Handoffs | Control transfers wholesale to a specialist | [support-desk](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/support-desk), [handoffs](/handoffs) | +| Handoffs | Control transfers wholesale to a specialist | [support-desk](https://github.com/LinuxDevil/lousho/tree/main/examples/support-desk), [handoffs](/handoffs) | ### Typed decisions ("System One" style) @@ -123,7 +123,7 @@ const triage = createAgent({ // and confidence < 0.7 routes to a human queue rather than acting. ``` -[coding-agent-workflows](https://github.com/LinuxDevil/agent-sdk/tree/main/examples/coding-agent-workflows) +[coding-agent-workflows](https://github.com/LinuxDevil/lousho/tree/main/examples/coding-agent-workflows) is the full example: a coding agent driven by a deterministic workflow whose branch points are typed decisions, where confidence below the floor exits `needs-human` instead of patching. diff --git a/providers.mdx b/providers.mdx index abc2d3b..e070c85 100644 --- a/providers.mdx +++ b/providers.mdx @@ -1,7 +1,7 @@ --- title: "Providers" --- -{/* Generated from docs/providers.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/providers.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} An `LLMProvider` is what an agent generates with. The SDK ships OpenAI, Anthropic, Google (Gemini), OpenRouter, Ollama and `pi` providers (each backed by an optional @@ -226,7 +226,7 @@ console.log(result.text); | ------ | ----------- | | `resolveProvider('p/model')` | Build a real provider from environment credentials. | | `LLMProviderRegistry` | Registry of provider factories: `create(type, config)`, `register(type, factory)`, `has(type)`. | -| `OpenAIProvider`, `AnthropicProvider`, `GoogleProvider`, `OllamaProvider`, `OpenRouterProvider`, `PiProvider` | The provider classes. See [examples/openrouter](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/openrouter) for OpenRouter-specific features. `PiProvider` is the `pi/...` adapter below. | +| `OpenAIProvider`, `AnthropicProvider`, `GoogleProvider`, `OllamaProvider`, `OpenRouterProvider`, `PiProvider` | The provider classes. See [examples/openrouter](https://github.com/LinuxDevil/lousho/blob/main/examples/openrouter) for OpenRouter-specific features. `PiProvider` is the `pi/...` adapter below. | | `createMockProvider()`, `MockLLMProvider` | Canned responses for demos; simulates a tool call when the message names a tool. | | `mockModel(script)` | Scripted, asserting provider for tests (from `/testing`); see [Testing](/testing). | | `withRetry()`, `withFallback()` | Retry transient failures with backoff and fall back to the next provider; see [Provider retries and fallback](/configuration#provider-retries-and-fallback). | diff --git a/queue-and-steer.mdx b/queue-and-steer.mdx index b92685c..a96a453 100644 --- a/queue-and-steer.mdx +++ b/queue-and-steer.mdx @@ -2,7 +2,7 @@ title: "Queued input and steering" sidebarTitle: "Queued input and steering" --- -{/* Generated from docs/queue-and-steer.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/queue-and-steer.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Add input to a run that is still going, or redirect it. Both are methods of the run handle that `agent.stream()` returns; see [Streaming](/streaming). diff --git a/quickstart.mdx b/quickstart.mdx index b16c869..59eb3a4 100644 --- a/quickstart.mdx +++ b/quickstart.mdx @@ -3,7 +3,7 @@ title: "Quick Start" sidebarTitle: "Quick start" description: "This page takes you from a one-line agent to tools, streaming, sessions, an approval and an offline test. Every snippet after the first runs as-is with no API key, on the SDK's built-in mock provider." --- -{/* Generated from docs/quick-start.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/quick-start.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## Start a new project @@ -242,7 +242,7 @@ have your own `LLMProvider` or want the mock for tests. `resolveProvider('/')` is the function `model` uses under the hood. This snippet uses OpenAI when `OPENAI_API_KEY` is set and falls back to the mock provider otherwise - the same pattern the runnable -[examples](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/README.md) use. +[examples](https://github.com/LinuxDevil/lousho/blob/main/examples/README.md) use. ```ts import { createAgent, createMockProvider, resolveProvider } from 'lousho'; @@ -314,4 +314,4 @@ lower-level options `createAgent()` does not, such as `temperature` and - [Configuration](/configuration) - every spec field, provider env var and CLI flag. - [Deployment](/deployment) - `lousho build` targets (Node server, Docker, Cloudflare Workers). - [API Overview](/api-overview) - the main exports and where to find full API reference. -- [Examples](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/README.md) - runnable example agents. +- [Examples](https://github.com/LinuxDevil/lousho/blob/main/examples/README.md) - runnable example agents. diff --git a/react.mdx b/react.mdx index 49e0d56..ef7e434 100644 --- a/react.mdx +++ b/react.mdx @@ -1,7 +1,7 @@ --- title: "React" --- -{/* Generated from docs/react.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/react.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `useLoushoAgent()` puts an agent behind a chat UI. It runs a turn, reads the [typed event stream](/streaming) as it arrives and turns it into diff --git a/reasoning.mdx b/reasoning.mdx index 2af4479..8ecc50b 100644 --- a/reasoning.mdx +++ b/reasoning.mdx @@ -1,7 +1,7 @@ --- title: "Reasoning" --- -{/* Generated from docs/reasoning.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/reasoning.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Reasoning models think before they answer. The `reasoning` option sets how much, and the run reports what the model thought separately from its answer: diff --git a/registry.mdx b/registry.mdx index c2e5b2e..78ab1f4 100644 --- a/registry.mdx +++ b/registry.mdx @@ -1,7 +1,7 @@ --- title: "Registry" --- -{/* Generated from docs/registry.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/registry.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `lousho add` installs a tool, skill, channel, schedule, memory slot or kit into an [agent directory](/agent-directories) from a registry. A `kit` item is a @@ -18,7 +18,7 @@ reads the default registry at `https://registry.lousho.com/index.json`, which is the `registry/dist/` folder of the SDK's repository served as static JSON. When that host cannot be reached, `lousho add` reads the same committed folder from GitHub instead: the release tag of the installed SDK version -(`https://raw.githubusercontent.com/LinuxDevil/agent-sdk//registry/dist/index.json`), +(`https://raw.githubusercontent.com/LinuxDevil/lousho//registry/dist/index.json`), then the same tag with a `v` prefix (older releases), then `main`. If GitHub cannot be reached either (offline, an outage, a version whose tag was never pushed), `lousho add` reads the copy of `registry/dist/` that ships inside diff --git a/runs.mdx b/runs.mdx index b34240a..e8c733b 100644 --- a/runs.mdx +++ b/runs.mdx @@ -3,7 +3,7 @@ title: "Runs: finish reasons, cancellation and parallel tools" sidebarTitle: "Runs and cancellation" description: "How a run ends, how to stop one, and how tool calls of one model turn run in parallel." --- -{/* Generated from docs/runs.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/runs.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## Finish reasons diff --git a/schedules.mdx b/schedules.mdx index 2a0bca3..34bf839 100644 --- a/schedules.mdx +++ b/schedules.mdx @@ -1,7 +1,7 @@ --- title: "Schedules" --- -{/* Generated from docs/schedules.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/schedules.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A **schedule** runs the agent on a cron expression, with nobody asking: a morning summary, a nightly cleanup. `defineSchedule()` declares one, an diff --git a/scripts/sync-pr-body.mjs b/scripts/sync-pr-body.mjs index 5e38bef..6fe6848 100644 --- a/scripts/sync-pr-body.mjs +++ b/scripts/sync-pr-body.mjs @@ -8,7 +8,7 @@ import { readFileSync, writeFileSync } from 'node:fs'; import { pathToFileURL } from 'node:url'; -const SDK_REPO = 'LinuxDevil/agent-sdk'; +const SDK_REPO = 'LinuxDevil/lousho'; /** Lines of the sync output that are warnings rather than the "wrote N pages" summary. */ export function parseSyncWarnings(output) { @@ -82,7 +82,7 @@ export function buildPullRequest({ sdkSha, syncOutput, syncExitCode, translation if (defaultToken) { out.push('', '> This pull request was opened with the default `GITHUB_TOKEN`, so the `Check docs` workflow does not run on it: run the checks locally or trigger them by hand before merging.'); } - out.push('', 'Source: LinuxDevil/agent-sdk#192.'); + out.push('', 'Source: LinuxDevil/lousho#192.'); return { title: `Sync with SDK docs at ${shortSha}`, body: out.join('\n') + '\n', draft }; } diff --git a/scripts/sync-sdk-docs.mjs b/scripts/sync-sdk-docs.mjs index 4c0f286..ccc9860 100644 --- a/scripts/sync-sdk-docs.mjs +++ b/scripts/sync-sdk-docs.mjs @@ -15,7 +15,7 @@ import { githubSlug, headings, mintSlug, slugList } from './anchors.mjs'; const siteRoot = resolve(dirname(fileURLToPath(import.meta.url)), '..'); const sdkRoot = resolve(process.argv[2] ?? join(siteRoot, '..', 'agent-sdk')); -const GITHUB = 'https://github.com/LinuxDevil/agent-sdk/blob/main/'; +const GITHUB = 'https://github.com/LinuxDevil/lousho/blob/main/'; // The SDK's docs/images/ (Studio screenshots) is mirrored to /images on this site. const SDK_IMAGES = 'docs/images/'; const SITE_IMAGES = 'images'; @@ -303,7 +303,7 @@ function convert(file, slug) { if (SIDEBAR_TITLES[slug]) front.push(`sidebarTitle: ${JSON.stringify(SIDEBAR_TITLES[slug])}`); if (description) front.push(`description: ${JSON.stringify(description)}`); front.push('---', ''); - const note = `{/* Generated from ${file} in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */}\n\n`; + const note = `{/* Generated from ${file} in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */}\n\n`; return `${front.join('\n')}${note}${converted.trimEnd()}\n`; } diff --git a/self-healing-paper.mdx b/self-healing-paper.mdx index f9861fb..c42d1cb 100644 --- a/self-healing-paper.mdx +++ b/self-healing-paper.mdx @@ -4,7 +4,7 @@ description: "A research-style report on Lousho's self-healing loop: how it reco --- -This page is a report, not a guide. It describes the design of the self-healing loop in the `lousho` SDK and the evidence collected up to 2026-11-09. For how to use it, read [Self-healing and self-improvement](/self-healing) and [Improving an agent](/improve). The SDK's [docs/improve.md](https://github.com/LinuxDevil/agent-sdk/blob/main/docs/improve.md) is the full reference for the loop. +This page is a report, not a guide. It describes the design of the self-healing loop in the `lousho` SDK and the evidence collected up to 2026-11-09. For how to use it, read [Self-healing and self-improvement](/self-healing) and [Improving an agent](/improve). The SDK's [docs/improve.md](https://github.com/LinuxDevil/lousho/blob/main/docs/improve.md) is the full reference for the loop. ## Abstract @@ -264,4 +264,4 @@ npx tsx examples/heal-benchmark/loop.ts --live --scenarios all --out ./m4-proof. npx tsx examples/heal-benchmark/loop.ts --report --out ./m4-proof.jsonl ``` -The report prints each bar with its counts and one verdict: GO, GO with misses, NO-GO, or INCOMPLETE. See the benchmark's [README](https://github.com/LinuxDevil/agent-sdk/blob/main/examples/heal-benchmark/README.md) for every flag. +The report prints each bar with its counts and one verdict: GO, GO with misses, NO-GO, or INCOMPLETE. See the benchmark's [README](https://github.com/LinuxDevil/lousho/blob/main/examples/heal-benchmark/README.md) for every flag. diff --git a/self-healing.mdx b/self-healing.mdx index 5c52ba6..e51d631 100644 --- a/self-healing.mdx +++ b/self-healing.mdx @@ -2,7 +2,7 @@ title: "Self-healing and self-improvement" description: "How Lousho recovers a failed run now (agent.recover(), a diagnosis retry, outage pauses, a progress guard) and proposes instruction and skill edits for later (improveAgent), which a person always approves." --- -{/* Generated from docs/self-healing.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/self-healing.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Lousho handles failures in two layers: diff --git a/sessions.mdx b/sessions.mdx index 6780042..8bb38a8 100644 --- a/sessions.mdx +++ b/sessions.mdx @@ -1,7 +1,7 @@ --- title: "Sessions" --- -{/* Generated from docs/sessions.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/sessions.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `agent.send(text)` is single-turn: every call starts from an empty history. A **session** is a multi-turn conversation: it keeps the transcript and passes it diff --git a/skills.mdx b/skills.mdx index 626a6d5..101dad3 100644 --- a/skills.mdx +++ b/skills.mdx @@ -1,7 +1,7 @@ --- title: "Skills" --- -{/* Generated from docs/skills.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/skills.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A skill is a named chunk of instructions (a markdown document) that the model loads **only when it needs it**. This is progressive disclosure: instead of diff --git a/stream-events.mdx b/stream-events.mdx index c955f9a..d7ad672 100644 --- a/stream-events.mdx +++ b/stream-events.mdx @@ -2,7 +2,7 @@ title: "Stream events" sidebarTitle: "Stream events" --- -{/* Generated from docs/stream-events.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/stream-events.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} The events `agent.stream()` emits, their order and their versioning. For how to consume a stream, see [Streaming](/streaming). diff --git a/streaming.mdx b/streaming.mdx index 4bafe12..43d185e 100644 --- a/streaming.mdx +++ b/streaming.mdx @@ -1,7 +1,7 @@ --- title: "Streaming" --- -{/* Generated from docs/streaming.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/streaming.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `agent.stream()` runs an agent exactly like `agent.send()` and reports the run as a stream of typed events: model text as it arrives, every tool call, diff --git a/structured-output.mdx b/structured-output.mdx index ff8b190..df2e743 100644 --- a/structured-output.mdx +++ b/structured-output.mdx @@ -1,7 +1,7 @@ --- title: "Structured output" --- -{/* Generated from docs/structured-output.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/structured-output.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Give an agent a zod schema as `output` and its final reply becomes a typed, validated object: `result.object`. The agent can still call tools first; only diff --git a/sub-agents.mdx b/sub-agents.mdx index 316e570..4373f33 100644 --- a/sub-agents.mdx +++ b/sub-agents.mdx @@ -1,7 +1,7 @@ --- title: "Sub-agents" --- -{/* Generated from docs/sub-agents.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/sub-agents.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A sub-agent is a separate agent - its own instructions, model, tools and skills - that a lead agent hands a self-contained task to. The sub-agent diff --git a/svelte.mdx b/svelte.mdx index 7323479..595fbd0 100644 --- a/svelte.mdx +++ b/svelte.mdx @@ -1,7 +1,7 @@ --- title: "Svelte" --- -{/* Generated from docs/svelte.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/svelte.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `loushoAgent()` for Svelte is the [React hook](/react) as a store: the same sources, the same state and the same actions, over the same diff --git a/testing.mdx b/testing.mdx index a077b40..87599ac 100644 --- a/testing.mdx +++ b/testing.mdx @@ -1,7 +1,7 @@ --- title: "Testing agents" --- -{/* Generated from docs/testing.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/testing.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `lousho/testing` ships `mockModel`: a scripted, deterministic `LLMProvider` for unit tests. You write down what the model should say on each diff --git a/tool-search.mdx b/tool-search.mdx index c97a065..dfad7bf 100644 --- a/tool-search.mdx +++ b/tool-search.mdx @@ -1,7 +1,7 @@ --- title: "Tool search" --- -{/* Generated from docs/tool-search.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/tool-search.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} An agent with many tools (several MCP servers, a large internal toolkit) sends every tool definition on every model call. That costs input tokens and, past diff --git a/tools.mdx b/tools.mdx index 82e67b1..91d95ae 100644 --- a/tools.mdx +++ b/tools.mdx @@ -1,7 +1,7 @@ --- title: "Tools" --- -{/* Generated from docs/tools.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/tools.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A tool is a typed function the model can call. Define one with `defineTool()`: the argument and result types are inferred from its zod `input`, the arguments diff --git a/triggers.mdx b/triggers.mdx index 7fe33d2..0956392 100644 --- a/triggers.mdx +++ b/triggers.mdx @@ -1,7 +1,7 @@ --- title: "Triggers" --- -{/* Generated from docs/triggers.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/triggers.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} A trigger adapter turns one wake-up mechanism into an agent run. The `lousho/triggers` subpath is the extension point for your own mechanism (`TriggerAdapter`, `TriggerRegistry`) plus a few helpers shared with channels and schedules: the `WebhookAuth` shapes, `verifySlackSignature()` and `parseCronExpression()`. The built-in webhook, Slack and cron adapters were removed; see [Migrating from triggers](/upgrading#migrating-from-triggers). diff --git a/troubleshooting.mdx b/troubleshooting.mdx index b9fac62..7157e2f 100644 --- a/troubleshooting.mdx +++ b/troubleshooting.mdx @@ -1,7 +1,7 @@ --- title: "Troubleshooting" --- -{/* Generated from docs/troubleshooting.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/troubleshooting.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} This page starts from what you see and points to the cause, the fix and the page that explains it. First run `npx lousho doctor` ([what it checks](/installation#troubleshooting-lousho-doctor)), @@ -38,7 +38,7 @@ environment (`LOUSHO_CONFIG_MISSING_PROVIDER`). - **Cause:** the quick start uses top-level `await`, so the file must be an ES module. - **Fix:** name the file `.mts`, or set `"type": "module"` in `package.json`. -- **More:** [Quickstart in the README](https://github.com/LinuxDevil/agent-sdk/blob/main/README.md#quickstart). +- **More:** [Quickstart in the README](https://github.com/LinuxDevil/lousho/blob/main/README.md#quickstart). ### The Node version is too old @@ -140,7 +140,7 @@ environment (`LOUSHO_CONFIG_MISSING_PROVIDER`). ## Getting help -Open an issue at [https://github.com/LinuxDevil/agent-sdk/issues](https://github.com/LinuxDevil/agent-sdk/issues). Include: +Open an issue at [https://github.com/LinuxDevil/lousho/issues](https://github.com/LinuxDevil/lousho/issues). Include: - the output of `npx lousho doctor --json`, - the error code (`LOUSHO_...`) and the full message, diff --git a/upgrading.mdx b/upgrading.mdx index 7cb5fdf..b40ef1a 100644 --- a/upgrading.mdx +++ b/upgrading.mdx @@ -2,7 +2,7 @@ title: "Upgrading to 1.0" sidebarTitle: "Upgrading to 1.0" --- -{/* Generated from docs/upgrading.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/upgrading.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## Who needs this page diff --git a/utilities.mdx b/utilities.mdx index 6f8d7e2..57f5a00 100644 --- a/utilities.mdx +++ b/utilities.mdx @@ -2,7 +2,7 @@ title: "Utilities" description: "Supporting helpers exported from the lousho/utils subpath: encryption and hashing, and file storage for attachments. None of them is needed to build an agent; they are here for the apps around one." --- -{/* Generated from docs/utilities.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/utilities.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} ## Encryption and hashing diff --git a/vue.mdx b/vue.mdx index 2ecb2ea..b4ed23e 100644 --- a/vue.mdx +++ b/vue.mdx @@ -1,7 +1,7 @@ --- title: "Vue" --- -{/* Generated from docs/vue.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/vue.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} `useLoushoAgent()` for Vue 3 is the [React hook](/react) as a composable: the same sources, the same state and the same actions, over the same diff --git a/workspace-tools.mdx b/workspace-tools.mdx index 8fe6602..862e48a 100644 --- a/workspace-tools.mdx +++ b/workspace-tools.mdx @@ -1,7 +1,7 @@ --- title: "Workspace tools" --- -{/* Generated from docs/workspace-tools.md in LinuxDevil/agent-sdk by scripts/sync-sdk-docs.mjs. Do not edit here. */} +{/* Generated from docs/workspace-tools.md in LinuxDevil/lousho by scripts/sync-sdk-docs.mjs. Do not edit here. */} Workspace tools give an agent a file system and a shell, so you can build coding agents in the style of Claude Code. The tools never use `node:fs` or