# Quipu

Quipu gives coding agents a local knowledge graph with rule-checked writes and time-aware queries. Use it when project facts need shared memory.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 8
- Author: intigraph
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/intigraph/quipu
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/mcp/quipu

## What it does

Quipu gives an agent a local store for facts, decisions, and work history. Each fact can retain its source, while the store tracks both the time of recording and the time when the fact held true. Queries use SPARQL 1.1, and Quipu exposes its own MCP server.

Writes pass through SHACL validation before they enter the graph. A rule violation is rejected with its reason, so invalid data is not stored. The Claude Code setup runs Quipu through Docker with `ghcr.io/scbrown/quipu:0.15.0`; Docker is required for this setup.

## Requirements
- Needs: Docker

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add quipu -- docker run -i --rm ghcr.io/scbrown/quipu:0.15.0
```

### Codex

**Platform CLI**
```bash
codex mcp add quipu -- docker run -i --rm ghcr.io/scbrown/quipu:0.15.0
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "quipu": {
            "command": "docker",
            "args": [
                "run",
                "-i",
                "--rm",
                "ghcr.io/scbrown/quipu:0.15.0"
            ]
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "quipu": {
            "type": "stdio",
            "command": "docker",
            "args": [
                "run",
                "-i",
                "--rm",
                "ghcr.io/scbrown/quipu:0.15.0"
            ]
        }
    }
}
```

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add quipu docker run -i --rm ghcr.io/scbrown/quipu:0.15.0
```

## When to use it
- store shared facts for coding agents
- validate graph writes with SHACL rules
- trace where a fact came from
- query facts by their valid time
- reject invalid data before it is stored

## Example prompts
- Save the decision about this module’s structure in Quipu and record its source.
- Check this new dependency fact against the SHACL rules and explain any rejection.
- Use Quipu to show which project facts were true on the requested date.

## Similar
- [open-ontologies](https://affpapa.org/en/nevibecode/plugins/open-ontologies.md) — plugin · ★ 1509
- [bgee-skill](https://affpapa.org/en/nevibecode/skills/bgee-skill.md) — skill · 13 installs
- [fedlex-mcp](https://affpapa.org/en/nevibecode/mcp/fedlex-mcp.md) — MCP server · ★ 7
- [Fedlex Connector](https://affpapa.org/en/nevibecode/mcp/fedlex-connector.md) — MCP server · ★ 7
- [mirofish-offline-simulation](https://affpapa.org/en/nevibecode/skills/mirofish-offline-simulation.md) — skill · 1400 installs
- [neo4j-document-import-skill](https://affpapa.org/en/nevibecode/skills/neo4j-document-import-skill.md) — skill · 923 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md