the-herald

Pass

Audited by Gen Agent Trust Hub on Jul 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes git commands (including git log, git tag, and git push) to manage repository history, versioning, and remote tags. This is the intended primary functionality for automating software releases.
  • [PROMPT_INJECTION]: An indirect prompt injection surface exists because the skill processes untrusted input from commit messages via git log to generate changelogs and release notes.
  • Ingestion points: Commit history retrieved via git log (referenced in SKILL.md).
  • Boundary markers: None present to isolate commit data from instructions.
  • Capability inventory: File system write access (CHANGELOG.md) and network operations (git push).
  • Sanitization: The skill translates technical commit subjects into user-facing language but does not perform explicit security filtering or instruction stripping from the input data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 8, 2026, 10:05 AM
Security Audit — agent-trust-hub — the-herald