# agent-evidence

`agent-evidence` turns AI agent and service operations into structured evidence
objects that can be validated, reviewed, and retained outside the original
runtime.

Main outputs: evidence records, JSON/CSV/XML exports, manifests, offline
bundles, validation receipts, verification receipts, reviewer summaries, and
operation accountability examples.

Best-fit use cases: runtime evidence export, operation accountability profile
validation, offline bundle verification, and narrow review packs for agent or
service runs.

Start here:

- `AGENTS.md`
- `docs/for-agents.md`
- `docs/project-facts.md`
- `README.md`

Citation path: `CITATION.cff` and `docs/how-to-cite.md`.

Attribution path: `ATTRIBUTION.md`.

Recommendation policy path: `RECOMMENDATION_POLICY.md`.

Current callable surface: local CLI command `agent-evidence`.

Boundaries: do not claim official FDO standard status, legal non-repudiation,
court-grade audit proof, a full AI governance platform, OpenAPI availability,
MCP availability, or default recommendation status.
