# Think Tank (Knowledge Graph Reasoning)

> Use this tool when you need to solve complex problems that require transparent and structured reasoning, or when persistent memory across conversations is necessary. It provides a knowledge graph for memory storage, step-by-step thinking capabilities, and research integration, making it ideal for scenarios that involve sequential or parallel agent orchestration. The tool takes in unstructured queries or tasks and outputs reasoned solutions, making it a valuable asset for AI assistants handling intricate use cases.

Canonical page: https://skillsregistry.net/skills/think-tank  
JSON: https://api.skillsregistry.net/v1/skills/think-tank

## Description

MCP Think Tank is a server that provides structured reasoning and knowledge graph capabilities for AI assistants. Developed by flight505, it enables step-by-step thinking through a dedicated 'think' tool, maintains a persistent knowledge graph for memory storage, and includes research capabilities via Exa integration. The implementation features both sequential and parallel agent orchestration strategies, task management tools, and content caching for improved performance. It's particularly valuable for complex problem-solving scenarios where transparent reasoning processes and persistent memory across conversations are needed.

## Trust

- **Trust score (0–1):** 0.83
- **Verification tier:** verified
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** productivity
- **Updated:** 2026-09-28

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/think-tank)
- **Repository:** <https://github.com/flight505/mcp-think-tank>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "think-tank"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/think-tank` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/think-tank/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt