MIT-licensed tools that make agent runtimes easier to connect, inspect, secure, and trust.
Secure Agent · Repositories · Documentation · Plugins
Secure Agent is an egress-inspection and secret-leak firewall for local AI agents on macOS. It correlates a sensitive-file read with the network egress that follows and flags the leak before a key leaves the machine.
Open building blocks for secure, local, and inspectable agent systems.
| Project | What it does |
|---|---|
| CAVI API Client | Gateway-agnostic TypeScript client — agent runtime infrastructure for typed HTTP, WebSocket, and SSE. |
| Bobby Browser | Alpha browser-automation runtime with authenticated, capability-scoped control across Rust, TypeScript, MCP, and CDP. |
| MLX Agent | Discovers, verifies, and wires local MLX-optimized models on Apple Silicon — Claude, Codex, Antigravity, OpenCode, and AgentSkills. |
| MLX Workbench | Loopback-local UI for the MLX Agent model lifecycle. |
| Companion for Claude | The Obsidian Community Store release for Claude knowledge workflows. |
| Obsidian Agent | Portable, CLI-powered Obsidian workflows across agent hosts. |
| MCP Eval | Evidence-driven MCP-server evaluation across discovery cost, schema guessability, error honesty, state recovery, and contention — turning agent friction into actionable findings. |
| Ableton MCP | Local-first Ableton Live control for MCP clients — Remote Script bridge, stdio MCP server, and CLI with confirm-to-execute mutations. |
| CAVI Plugins | Host-neutral plugin catalog across Claude, Codex, Gemini, OpenCode, and AgentSkills. |
| Antigravity for OpenClaw | OpenClaw provider plugin for Google's Antigravity CLI. |
- Agent security — tying a sensitive-file read to the network egress that follows.
- Runtime interoperability — connecting gateways and providers through typed boundaries.
- Local-first workflows — running models and knowledge tools on hardware you control.
- MCP reliability — measuring discovery cost, schema guessability, error honesty, state recovery, and contention.
Every public CAVI-AI project ships with its source and provenance. Read the code, test the assumptions, adapt the tools, and contribute what you learn. More work is in development — we publish it when there is working software and evidence worth examining.
cavi-ai.xyz · all repositories · plugin catalog
MIT licensed · Source published · Provenance attached
