skills/webriq/claude-skills/release/Gen Agent Trust Hub

release

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Automates the release process using standard git operations such as git describe, git tag, git commit, and git push, as well as the GitHub CLI gh release create command for release management.\n- [EXTERNAL_DOWNLOADS]: Includes references to official GitHub repositories from trusted organizations such as Vercel Labs and Supabase for optional best-practice guidelines.\n- [PROMPT_INJECTION]: The skill processes untrusted input from local TASKS.md and task documentation files to generate release notes and changelog entries, which constitutes an indirect prompt injection surface.\n
  • Ingestion points: Reads content from the TASKS.md file and individual markdown task documents within the repository.\n
  • Boundary markers: No specific delimiters or safety instructions are provided to the agent to distinguish between its own logic and the ingested task content.\n
  • Capability inventory: The skill has the ability to write to the local filesystem (CHANGELOG.md, TASKS.md), perform git commits/pushes, and create GitHub releases via CLI.\n
  • Sanitization: No explicit sanitization or filtering is performed on the ingested task descriptions before they are used in automated release notes.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 02:05 AM