progress-contract
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is project management and reporting within GitHub repositories. It uses standard integrations for monitoring issues and pull requests.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources such as GitHub issue descriptions and comments ('Re-read the original ask'). While this presents an attack surface for indirect injection, the risk is mitigated by the skill's defined scope; the '/lever status' command is explicitly read-only and does not perform destructive actions like merging or scheduling based on untrusted input.
- [DATA_EXPOSURE_EXFILTRATION]: The skill references a local configuration file ('~/.copilot/lever.yml'). This is used for repository discovery and status tracking. No evidence of credential exfiltration or unauthorized network operations to non-trusted domains was found.
- [COMMAND_EXECUTION]: The skill references internal commands like '/lever status' and '/lever deliver'. These are specific to the 'lever' toolset and do not involve arbitrary shell execution of user-supplied strings.
Audit Metadata