Loading connector details…
Loading connector details…
Choose a unique username to continue using AgentHotspot
by cgize • Legal & Compliance
An MCP server that implements Anthropic's "think" tool to give Claude a structured scratchpad for improved multi-step reasoning and policy-compliant decision making.
A dedicated scratchpad to break down complex problems, list applicable rules, and plan multi-step actions before executing tools.
Verify and iterate over tool outputs and ensure planned actions comply with detailed policies and constraints.
Persistent thinking logs and analytics (retrieve, clear, and analyze thoughts) to audit or improve reasoning strategies.
This server provides a Model Context Protocol (MCP) implementation of Anthropic's "think" tool, exposing endpoints to record, retrieve, clear, and analyze structured reasoning ('think', 'get_thoughts', 'clear_thoughts', 'get_thought_stats'). It is designed to boost Claude's performance on complex, multi-step tasks by offering a dedicated space to plan, check rules, iterate over tool outputs, and verify policy adherence. Installation options include a Smithery install command or npm package, and configuration snippets for integrating with Claude Desktop are provided. The project is open-source (MIT) and intended for integration into agent tool-chains that require careful reasoning management.
Use this tool to think about something. It will not obtain new information or change anything, but just append the thought to the log. Use it when complex reasoning or cache memory is needed, especially during long chains of tool calls, policy adherence scenarios, or sequential decision making.
Retrieve all thoughts recorded in the current session to review your reasoning process.
Clear all thoughts recorded in the current session. Use this to start fresh if the thinking process needs to be reset.
Get statistics about the thoughts recorded in the current session to analyze your thinking process.
Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.