github-code-search
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a guide for using official, platform-provided GitHub MCP tools (such as
search_codeandget_file_contents) for their intended purpose of researching implementation patterns. - [DATA_EXFILTRATION]: No unauthorized data exfiltration patterns were detected. The skill interacts with public GitHub data using authenticated tools, and there are no instructions to access or transmit sensitive local environment variables or private credentials.
- [INDIRECT_PROMPT_INJECTION]: The skill involves ingesting untrusted external data (GitHub source code), which is an inherent surface for indirect prompt injection. However, the instructions explicitly mitigate this by advising the agent to treat retrieved code as reference material only and to adapt it to local conventions rather than executing or copying it blindly.
- [PROMPT_INJECTION]: No attempts to override system instructions, bypass safety filters, or extract system prompts were found.
Audit Metadata