Lp3
Medium
- Category
- MCP Least Privilege
- Confidence
- 70% confidence
- Finding
- Without declared permissions the skill's intent is opaque and cannot be validated.
Security checks across malware telemetry and agentic risk
This skill locally analyzes GitHub Actions JSON exports to identify flaky CI steps and does not show hidden network, credential, persistence, or destructive behavior.
Install if you are comfortable letting the skill read the GitHub Actions JSON files matched by RUN_GLOB. Keep RUN_GLOB narrow, because those exports may include repository names, branches, run IDs, job and step names, and GitHub URLs.
64/64 vendors flagged this skill as clean.