orchestrate-projects-by-jean
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a Python script (scripts/jean_ops.py) to interact with local development tools via subprocess.Popen. These executions are conducted securely using a list of arguments (avoiding shell interpretation) and are restricted to a hardcoded allowlist of read-only MCP tools.- [DATA_EXPOSURE]: The skill accesses project-specific configuration files and agent session history. To mitigate risks, the execution script implements automatic redaction of sensitive patterns such as API tokens and passwords, and enforces string length limits on all data returned to the agent.- [EXTERNAL_DOWNLOADS]: The skill relies on local binaries (mcp2cli, jean) and project dependencies. These are standard components for the platform's orchestration workflow and are managed through the author's own marketplace/repositories.
Audit Metadata