pr-lifecycle

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references supporting modules from the vendor's 'dev-skills' repository, such as 'ci-triage' and 'issue-spec', which provide shared functionality for the development lifecycle.
  • [PROMPT_INJECTION]: An indirect prompt injection surface exists as the skill ingests content from external sources to update repository state.
  • Ingestion points: Reads PR bodies and issue descriptions via mcp__github__pull_request_read and mcp__github__search_issues as defined in SKILL.md.
  • Boundary markers: The skill does not specify delimiters to isolate external data from its operational instructions.
  • Capability inventory: The skill has the capability to write to the repository using mcp__github__issue_write, mcp__github__update_pull_request, and mcp__github__add_reply_to_pull_request_comment in SKILL.md.
  • Sanitization: No processes for sanitizing or validating the content retrieved from GitHub are described before the data is used to modify issues or PRs.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 10:41 PM
Security Audit — agent-trust-hub — pr-lifecycle