mcphub-operator
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run local bash and Python scripts (scripts/preflight.sh, scripts/check.py) to validate the environment and settings. These scripts are intended for local repository maintenance and use standard subprocess calls without shell evaluation for external input.
- [DATA_EXPOSURE]: The doctor.py script checks for the existence of the MCPHUB_BEARER_TOKEN environment variable to verify the hub's configuration. It does not print or transmit the token value, only reporting its presence status to the agent, adhering to safe credential management practices.
- [DYNAMIC_EXECUTION]: Python scripts in the repository use subprocess.run to execute local validation utilities and manipulate sys.path to import internal modules from the repo root. This behavior is localized to the user's project structure and is consistent with the skill's purpose as a developer operator.
- [INDIRECT_PROMPT_INJECTION]: The skill processes internal JSON configuration files (mcp-registry.json). It mitigates potential injection risks by providing the agent with clear scoping instructions for workflow groups, limiting the tool surface area and context bloat. Ingestion points: config/mcp-registry.json. Boundary markers: SKILL.md definitions for the harness group. Capability inventory: subprocess.run in check.py and doctor.py. Sanitization: Standard JSON parsing via the json module.
Audit Metadata