Agent Evidence

Agent Evidence is a portable draft standard for agent evidence, provenance, verification, review, replay, redaction, and audit export.

It interoperates with runtime, UI, knowledge, artifact, telemetry, policy, tool, task, and peer-agent systems without taking ownership of those systems. Runtime produces execution facts. UI renders evidence. Knowledge and retrieval systems provide source material. Artifact services own bytes and versions. Evidence packages the trust graph that links outcomes to support, contradiction, review, replay, privacy, and export facts.

What v0.1.0 defines

• Evidence Pack as the portable container for scoped evidence.
• Claim maps that connect assertions to supporting, contradicting, qualifying, and background evidence.
• Source maps with selectors, snippets, retrieval metadata, omissions, trust, freshness, and counter-evidence.
• Provenance chains for entities, activities, agents, tools, models, artifacts, humans, and peer systems.
• Verification results and review verdicts as separate facts.
• Replay cases, redaction records, privacy metadata, telemetry correlation, and export manifests.
• Public JSON Schemas and LLM-friendly llms.txt / llms-full.txt entrypoints.

Documentation

• Specification
• Evidence model
• Evidence pack
• Claim map
• Source map
• Provenance chain
• Verification and review
• Replay case
• JSON Schemas
• Research sources
• 中文规范

LLM entrypoints

• llms.txt: concise navigation index for AI clients.
• llms-full.txt: concatenated current English documentation with source URLs.
• llm.txt and llm-full.txt: compatibility aliases.

Related Agent standards

• Agent Knowledge - source-grounded knowledge packs.
• Agent UI - interaction surfaces for agent products.
• Agent Runtime - execution facts, controls, tasks, tools, and recovery.
• Agent Evidence - evidence, provenance, verification, review, replay, and export.
• Agent Policy - policy decisions, approvals, permissions, risk, retention, waivers, and traces.
• Agent Artifact - durable deliverables, versions, parts, previews, exports, and handoff packages.
• Agent Tool - tool declarations, surfaces, invocations, progress, results, permissions, and audit refs.
• Agent Context - context surfaces, items, source refs, selection, budgets, assembly, injection, compaction, and missing-context facts.

See the Agent standards ecosystem page for the mutual-link map and future standard candidates.

Local development

npm install
npm run dev

Build

npm run build

The static site is generated at docs/.vitepress/dist.