common-store-changelog

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run git log <last-tag>..HEAD --oneline in SKILL.md. This is a standard and necessary operation for the skill's primary purpose of gathering commit history to generate changelogs.
  • [DATA_EXPOSURE]: No sensitive file access (such as SSH keys or environment variables) or credential exfiltration patterns were detected. The skill only interacts with git commit data for processing.
  • [PROMPT_INJECTION]: The instructions do not contain attempts to bypass safety filters, override agent behavior, or extract system prompts.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes potentially untrusted input in the form of git commit messages, the output is restricted to user-facing release notes intended for human review. There is no automated execution of the generated content that would pose a risk. The ingestion point is the git log output, and while explicit boundary markers are absent, the transformation logic acts as a natural filter.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 07:08 PM
Security Audit — agent-trust-hub — common-store-changelog