Skip to content
View hernancapucci's full-sized avatar
💭
Open-source specifications.
💭
Open-source specifications.

Organizations

@digital-existence-layer

Block or report hernancapucci

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
hernancapucci/README.md

Hernán Capucci

Real estate broker in Rosario, Argentina, and builder of digital products. I also do independent research on how AI agents declare who they are, what they are allowed to do and on whose behalf.

Corredor inmobiliario en Rosario y creador de productos digitales. Investigo de forma independiente la gobernanza de agentes de IA.


Research

Independent, open work. Archived with DOIs on Zenodo for community review — not peer-reviewed.

  • Agent Manifest — an open declarative specification (v1.0) for AI agents to publish their identity, capabilities and operational boundaries before interacting. Specification, JSON Schema, a command-line validator, a reading client and a public registry live in the agent-manifest organization. DOI: 10.5281/zenodo.18833955
  • Agentic Governance & Trust Stack (AGTS) — an independent working paper (v0.2.1, 2026), separate from the Agent Manifest specification. A five-layer framework for trust and governance around autonomous economic agents: declaration, identity, policy, runtime enforcement and decision forensics. DOI: 10.5281/zenodo.21347422
  • The Pre-Execution Identity Gap in AI Agents — working paper (2026) proposing a minimal declarative layer before an agent acts. DOI: 10.5281/zenodo.19432967
  • ∈ Principle — a philosophical and institutional proposal on public authorship and responsibility in the age of AI, also published as a book in Spanish and English. DOI: 10.5281/zenodo.18965668

Other works, such as Boundary Handshake and a comparative study on boundary declaration, are listed on ORCID.

Tools

  • skill-static-review — prompts and checklists to statically review third-party AI skills before installing them. It is not a scanner and does not certify safety.
  • llm-waste-audit — an early MVP that detects avoidable LLM waste, such as cache breakers and context bloat.

Products

  • Existence Compiler — an early-stage product that scans a public URL and reports which discovery conditions for search engines and AI systems it could evaluate, with its coverage. What it cannot determine is not counted as a failure.
  • OwnSite Studio — the custom website design and development studio I founded in 2026.

Book


Elsewhere: LinkedIn · X · ORCID · Medium

In academic citations and records I appear as Hernán Alfredo Capucci.

Pinned Loading

  1. agent-manifest/agent-manifest agent-manifest/agent-manifest Public

    Declarative specification for AI agents to publish identity, capabilities and operational boundaries before interaction.

    JavaScript 2

  2. agent-manifest/agentic-governance-trust-stack agent-manifest/agentic-governance-trust-stack Public

    Infrastructure research on declaration, identity, policy, runtime control, and decision forensics for autonomous agents.

  3. skill-static-review skill-static-review Public

    Static review checklists and prompts for inspecting third-party skills before installation.

  4. llm-waste-audit llm-waste-audit Public

    Detect avoidable LLM waste: cache breakers, context bloat, and inefficient prompt design

    Python

  5. el-nuevo-creador-tecnico el-nuevo-creador-tecnico Public

    Official description of 'El nuevo creador técnico' — a digital reference book by Hernán Capucci for non-technical professionals who want to understand digital systems and AI without becoming progra…