patent-mcp-setup
Pass
Audited by Gen Agent Trust Hub on Apr 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs standard setup operations for an MCP (Model Context Protocol) server. This includes creating a virtual environment, installing local package dependencies, and registering the server in the agent's configuration files (~/.claude/settings.json and ~/.config/opencode/opencode.json).
- [COMMAND_EXECUTION]: The installation workflow involves executing local shell commands and Python scripts for environment verification and package installation. These commands are restricted to the local environment and are necessary for the setup process.
- [DATA_EXFILTRATION]: While the skill communicates with the external domain 'plus.kipris.or.kr', these interactions are the intended core functionality (fetching patent data via the official KIPRIS API). The code includes sensitive data masking in logs to prevent the leakage of API keys.
- [DYNAMIC_EXECUTION]: The skill utilizes dynamic module loading (importlib.import_module) within its internal registry to discover and load tool classes. This is a standard software pattern used here for modularity and does not involve executing untrusted or remote code.
- [EXTERNAL_DOWNLOADS]: The skill instructions and install script fetch dependencies from standard package registries (PyPI) via 'pip' or 'uv', which is typical for Python-based agent skills.
Audit Metadata