Evidence Guard

Audit any AI-generated output for unsupported claims, then verify every factual and technical assertion against its real source before it ships.

Install
cmdop skills install agensi-evidence-guard

Evidence Guard is a skill that applies regulated-industry evidence standards to AI-generated output. It targets a specific failure mode: AI agents produce confident, fluent text that can contain wrong version numbers, references to deprecated APIs, performance figures with no backing benchmark, and documentation that has drifted away from the actual codebase. A single model reviewing its own work tends to approve the same errors it introduced.

Evidence Guard addresses this with a structured Claims-QC pass. It extracts every factual, technical, and quantitative claim from a piece of output, classifies each claim by type, verifies it against the repository or a citable source, grades the strength of the supporting evidence, and flags specific risk patterns including version mismatches and doc-versus-code drift. No output passes the verdict gate until every critical claim is traceable to a real source.

The skill is designed to run before an agent ships documentation, READMEs, PR descriptions, API references, or changelogs. The end product is a compact, audit-ready Verification Note suitable for dropping into a pull request or a documentation review workflow. The approach draws on evidence disciplines from medical and scientific publishing, specifically the rigor used in MLR and peer review processes, and applies them to everyday developer-facing agent output. There are no environment variables required and no transport is specified.

Use cases

  • Verify that all version numbers in a generated README match actual published releases before merging
  • Check AI-written API reference documentation for deprecated endpoint references
  • Validate performance claims in a generated changelog against traceable benchmark sources
  • Catch doc-vs-code drift in PR descriptions before a pull request is reviewed
  • Produce an audit-ready Verification Note for regulated-industry documentation workflows
  • Run a Claims-QC pass on agent-generated technical content to flag unsupported quantitative assertions

When to use it

  • When an AI agent is generating documentation, READMEs, or API references that will be published or reviewed
  • When a workflow requires traceable, audit-ready evidence for every technical claim
  • When version numbers, deprecation status, or performance figures in generated content need to be verified against a real source
  • When doc-vs-code drift is a known risk in a rapidly changing codebase

When not to use it

  • When the output contains no factual or technical claims that require source verification
  • When no repository or citable source is available to verify claims against
  • When the task requires real-time data retrieval rather than static claim verification
  • When a transport-specific MCP server integration is required, as no transport is specified for this skill