m42sd-skill
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill follows security best practices for interacting with enterprise APIs.
- [DATA_EXFILTRATION]: While the skill handles sensitive API tokens, it implements secure storage practices (0600 file permissions) and only communicates with the user-defined base URL.
- [INDIRECT_PROMPT_INJECTION]: The skill explicitly instructs the agent to treat all fetched data as untrusted. The underlying Python script uses proper escaping (html.escape and asql_quote) when processing data to be sent back to the Matrix42 API or used in queries, mitigating injection risks.
- [COMMAND_EXECUTION]: Static analysis detected subprocess.run in the test suite (tests/test_m42_live.py). Review of the code confirms this is used legitimately for integration testing to execute the skill's own CLI script with controlled arguments, posing no risk to the operational environment.
Audit Metadata