smell
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code from the project being analyzed, which could contain malicious instructions designed to subvert the agent's behavior.
- Ingestion points: The agent scans project files, specific directories, or git diffs using find, grep, and an Explore subagent as described in SKILL.md.
- Boundary markers: The instructions lack explicit delimiters to isolate the code data from the agent's task instructions.
- Capability inventory: The skill has the ability to read files, invoke a subagent for exploration, and write markdown reports to the tasks/ directory (SKILL.md).
- Sanitization: There is no evidence of content filtering or sanitization of the code before it is passed to the LLM for analysis.
- [COMMAND_EXECUTION]: The skill executes shell commands to facilitate codebase discovery.
- Evidence: The job description in SKILL.md specifies using find and grep to gather evidence for the architectural report.
Audit Metadata