generate-release-notes

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local git commands (git log, git describe, git remote) and a bash script included in the skill's directory (scripts/parse-commits.sh) to retrieve and process repository history.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests untrusted data from git commit messages which are then presented to the agent for summarization.
  • Ingestion points: Git repository history retrieved via git log and processed by scripts/parse-commits.sh.
  • Boundary markers: The script uses a specific output format with defined fields (COMMIT:, TYPE:, etc.) and a custom delimiter (---COMMIT-BOUNDARY---) to structure the data for the agent.
  • Capability inventory: The skill can execute local git commands and write to files (RELEASE_NOTES.md, CHANGELOG.md).
  • Sanitization: The parse-commits.sh script employs regular expressions to extract specific fields from the commit messages, providing a layer of structure before the agent processes the content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 04:39 AM
Security Audit — agent-trust-hub — generate-release-notes