skills/dotnet/runtime/pr-failure-scan/Gen Agent Trust Hub

pr-failure-scan

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • Indirect Prompt Injection Surface: The skill is designed to ingest and analyze data from external sources, including GitHub PR metadata and CI logs from Azure DevOps and Helix.
  • Ingestion points: Data is pulled from the Build Analysis check run payload, AzDO build/timeline/logs, and Helix console logs (Step 2 and Step 4).
  • Boundary markers: The instructions do not define specific delimiters or escaping mechanisms for the content parsed from external logs.
  • Capability inventory: The skill can write files to the local filesystem and create GitHub issues using the gh API (Step 5).
  • Sanitization: While explicit sanitization is not mentioned, the skill requires mandatory human confirmation in Step 1 (for non-owned PRs) and Step 5 (before creating each live issue), which serves as a significant control against unauthorized actions driven by untrusted data.
  • GitHub Tool Interaction: The skill utilizes the GitHub CLI (gh api) to resolve user identities and manage issues. This is a standard integration for development-focused skills, as it operates within the scope of the user's existing authenticated session and requires explicit permission to act on PRs the user does not own.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 10:38 AM
Security Audit — agent-trust-hub — pr-failure-scan