{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_eq9zjmuuenfv","handle":"iflow-mcp-beverm2391-chain-of-thought-mcp","url":"https://wellknown.network/agents/iflow-mcp-beverm2391-chain-of-thought-mcp","links":{"self":"https://wellknown.network/agents/iflow-mcp-beverm2391-chain-of-thought-mcp/record.json","html":"https://wellknown.network/agents/iflow-mcp-beverm2391-chain-of-thought-mcp","markdown":"https://wellknown.network/agents/iflow-mcp-beverm2391-chain-of-thought-mcp/record.md","api":"https://wellknown.network/api/v1/agents/iflow-mcp-beverm2391-chain-of-thought-mcp","status":"https://wellknown.network/api/v1/agents/iflow-mcp-beverm2391-chain-of-thought-mcp/status","claim":"https://wellknown.network/agents/iflow-mcp-beverm2391-chain-of-thought-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/iflow-mcp-beverm2391-chain-of-thought-mcp/claim.json","badge":"https://wellknown.network/agents/iflow-mcp-beverm2391-chain-of-thought-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/iflow-mcp-beverm2391-chain-of-thought-mcp/history","tools":"https://wellknown.network/api/v1/agents/iflow-mcp-beverm2391-chain-of-thought-mcp/tools"},"ard":{"identifier":"urn:air::server:iflow-mcp-beverm2391-chain-of-thought-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"iflow-mcp_beverm2391-chain-of-thought-mcp","summary":"A MCP server that uses Groq's API to call LLMs which expose raw chain-of-thought tokens.","description":"# Chain of Thought MCP Server\n\nAnthropic's recent article [\"The \"think\" tool: Enabling Claude to stop and think in complex tool use situations\"](https://www.anthropic.com/engineering/claude-think-tool) shows that using an external `think` tool notably increases performance on [SWE Bench](https://www.swebench.com/).\n\nThis MCP Server uses Groq's API to call LLMs which expose raw chain-of-thought tokens from Qwen's [qwq model](https://qwenlm.github.io/blog/qwq-32b/).\n\n## Installation \n\n1. Clone this repository to your local machine.\n2. Run ```uv sync``` to install depencies\n3. Get a Groq API key from [here](https://console.groq.com/keys).\n4. Update your mcp configuration with:\n\n```json\n\"mcpServers\": {\n  \"chain_of_thought\": {\n    \"command\": \"uv\",\n    \"args\": [\n        \"--directory\",\n        \"path/to/cot-mcp-server\",\n        \"run\",\n        \"src/server.py\"\n      ],\n      \"env\": {\n        \"GROQ_API_KEY\": \"your-groq-api-key\"\n      }\n    }\n}\n```\n\nThe path should be the local path to this repository. You can get this easily by running `pwd` in the terminal from the root of the repository.\n\n## Instructing The AI To Use This MCP Server\n\nI personally prefer the agent call this tool on every request to increase performance. I add this to my rules for the agent:\n\n```xml\n<IMPORTANT>\n<when_to_use_tool>\nYou should call the mcp chain_of_thought tool every time you talk to the user, which generates a chain-of-thought stream which you will use to complete the user's request.\n</when_to_use_tool>\n\nBefore taking any action or responding to the user use the chain of thought tool as a scratchpad to:\n- List the specific rules that apply to the current request\n- Check if all required information is collected\n- Verify that the planned action complies with all policies\n- Iterate over tool results for correctness \n\nHere are some examples of what to iterate over inside the think tool:\n<cot_tool_example_1>\nUser wants to cancel flight ABC123\n- Need to verify: user ID, reservation ID, reason\n- Check…","publisher":null,"homepage":null,"repository":null,"version":"0.1.0","license":null,"protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:iflow-mcp_beverm2391-chain-of-thought-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","summary":"pypi","version":"pypi","description":"pypi"}},"derived":{"capabilities":[{"slug":"knowledge.reasoning","name":"Reasoning & Planning","confidence":1,"provenance":"derived"},{"slug":"dev.version-control","name":"Version Control","confidence":0.825,"provenance":"derived"}],"categories":["dev","knowledge"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"iflow-mcp_beverm2391-chain-of-thought-mcp","registry":"pypi","observedAt":"2026-11-09T07:25:38.512Z","publishedAt":"2026-02-06T17:29:52.642067Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"iflow-mcp_beverm2391-chain-of-thought-mcp","url":"https://pypi.org/project/iflow-mcp_beverm2391-chain-of-thought-mcp/","firstSeenAt":"2026-09-09T18:20:44.294Z","fetchedAt":"2026-11-09T07:24:40.594Z","normalizedAt":"2026-11-09T07:24:40.594Z"}]},"firstSeenAt":"2026-09-09T18:20:44.294Z","updatedAt":"2026-11-09T07:25:38.512Z"}