launchdarkly-flag-cleanup
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious instructions, obfuscation techniques, or unauthorized security risks were found. The skill's behavior is consistent with its stated purpose of cleaning up feature flags.\n- [EXTERNAL_DOWNLOADS]: The skill requires the '@launchdarkly/mcp-server', which is a legitimate vendor resource used to securely communicate with the LaunchDarkly API.\n- [COMMAND_EXECUTION]: Instructions guide the agent to perform standard development tasks such as searching code with 'grep' and running local project verification scripts (build/lint).\n- [DATA_EXFILTRATION]: Data access is limited to fetching flag configurations from the official LaunchDarkly API and scanning the local codebase for flag keys, which are necessary steps for the cleanup process.
Audit Metadata