prepare
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes local codebase files (such as configuration files and source code) to identify sensitive information and project structure. This creates a surface for indirect prompt injection if malicious instructions were embedded in the scanned files.
- Ingestion points: The skill reads ecosystem files (e.g.,
package.json,pyproject.toml),.gitignorefiles, and source code comments during the Discovery (Phase 1), Secrets Audit (Phase 2), and Metadata (Phase 7) phases. - Boundary markers: The instructions do not specify the use of clear delimiters or 'ignore' instructions when the agent reads external file content.
- Capability inventory: The skill is authorized to perform file writes (documentation, licenses, configuration), execute
gitcommands (commit, removal from cache), and call related internal skills (git-repo-prep:review). - Sanitization: No explicit sanitization or escaping of data retrieved from the codebase is mentioned before it is used to draft documentation or commit messages. However, the mandatory 'AskUserQuestion' step for every modification serves as a primary human-in-the-loop control.
Audit Metadata