deployment
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain malicious instructions aimed at influencing agent behavior.
- Ingestion points: The skill ingests external project code and developer-provided deployment preferences as specified in the 'Input' and 'Steps' sections of
SKILL.md. - Boundary markers: There are no explicit delimiters or specific 'ignore' instructions provided to separate these external inputs from the agent's operational instructions.
- Capability inventory: The skill is tasked with writing documentation to the filesystem and executing git commands for commits, as defined in the 'Steps' and 'Output' sections.
- Sanitization: No automated sanitization or filtering of input content is specified, although a 'Review Checklist' is included to manually verify that no sensitive information is hardcoded in environment variables.
Audit Metadata