feature-discovery
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill focuses on documentation and research tasks without performing dangerous operations.
- [DATA_EXPOSURE_AND_EXFILTRATION]: While the skill reads local files and extracts metadata (Step 1 and Step 3), it does not perform any network operations or send data externally. The operations are confined to the local project environment.
- [PROMPT_INJECTION]: The skill includes clear semantic boundaries using tags like
<what_you_do_not_do>which explicitly prohibit the agent from making technical decisions or writing code, mitigating risks from malicious input data. - [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or privilege escalation attempts detected in the instructions.
Audit Metadata