impact-check
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by ingesting untrusted data from codebase files and using it to generate markdown reports.
- Ingestion points: The skill reads arbitrary files across the codebase in the 'Steps' section to identify symbol references.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to prevent it from obeying instructions potentially embedded in the source code being analyzed.
- Capability inventory: The skill performs file-write operations to the
llmwiki/directory, including creating new draft files and appending to indices and logs. - Sanitization: No sanitization or validation of the extracted codebase content is performed before it is interpolated into the wiki reports.
Audit Metadata