changelog
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard
gitshell commands (git describe,git log) to extract version tags and commit history for analysis. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from git commit logs which could contain malicious instructions.
- Ingestion points: Commit messages and bodies are ingested via
git loginSKILL.md(Step 1). - Boundary markers: The workflow lacks explicit delimiters or instructions to the agent to disregard potential commands embedded within the commit history.
- Capability inventory: The skill allows the agent to write files to the local repository (
CHANGELOG.md,RELEASES.md, and migration guides) and can trigger additional actions through collaborating skills likegit-commit. - Sanitization: There is no evidence of filtering or sanitization of the commit data before it is processed by the AI agent.
Audit Metadata