A Chain of Thought MCP server that exposes raw reasoning tokens from large language models through Groq's API. The server captures and streams the model's thinking process by intercepting content within <think> tags, allowing AI assistants to reveal their step-by-step reasoning before providing final answers. Built with FastMCP and supporting models like Qwen and DeepSeek, it enables more transparent AI interactions where users can see how the model arrives at conclusions rather than just receiving the final output.
Cognium trust score
91%
Tier
Scanned
Composite of vulnerability cleanliness, spec conformance, provenance, stability, and usage signals — scanned and weighted by Cognium. Human and agent signals are tracked separately.
Last scanned 2026-09-19.
Returns 7 tools: search_skills, get_skill, list_leaderboard, get_trust_breakdown, resolve_composition, plus the ChatGPT-connector search and fetch. Every tool is annotated read-only.
Resolve this skill directly via MCP tools/call get_skill.