changelog-composer
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill workflow requires the agent to execute shell commands using standard tools, specifically
gitandgh(GitHub CLI), to fetch repository metadata and history (seeSKILL.md, Workflow Phase 1). - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) due to its processing of external, untrusted text data.
- Ingestion points: Repository data including commit messages and PR descriptions are fetched via
git logandgh pr listand passed to the agent for analysis (SKILL.md). - Boundary markers: The instructions do not implement boundary markers or specific guidance to ignore potential instructions embedded within the git history or PR content.
- Capability inventory: The skill leverages CLI tools to read repository history and structure changes into human-readable formats.
- Sanitization: There are no explicit sanitization steps defined to filter malicious payloads or directives that might be present in PR descriptions or commit bodies.
Audit Metadata