The Agent Skills Directory

[GENERAL] (SAFE): The skill is well-structured and focuses on enhancing reasoning capabilities. It is a tool-independent conversion of the verified Anthropic Sequential Thinking MCP server.
[COMMAND_EXECUTION] (SAFE): The included scripts (process-thought.js, format-thought.js) are local utility tools. They do not execute arbitrary shell commands or perform unsafe operations. They are designed for structured data processing and local state management.
[DATA_EXPOSURE] (SAFE): Thought history is persisted in a local JSON file (.thought-history.json). No sensitive system information or credentials are accessed or stored.
[EXTERNAL_DOWNLOADS] (SAFE): The project uses jest for testing, which is a standard and trusted dependency from a verified organization (Meta).

sequential-thinking