# DecisionPen Context Engine

The DecisionPen Context Engine MCP server for Claude Code gives agents access to DecisionPen’s architecture graph, decisions, and documents. It supports architecture research and sends proposed ADRs for human review.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: DecisionPen
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/DecisionPen/decisionpen-mcp
- Updated: 2026-11-15
- Security check: automated, 2026-11-17 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/decisionpen-context-engine

## What it does

The server lets an agent look up graph entities and their links, then read related evidence, decision records, and shared standards. If an S3 repository is connected, the agent can search its permitted documents too.

Agents can send documents or enrichment proposals to the Inbox. A human must approve them before they become confirmed entries on the architecture map. Set `DP_MCP_MODE=propose` or `persist_approved` to submit content; the default `cite_only` mode supports reading and citations without submissions.

Connecting to the hosted server requires an Edge API token. Its endpoint is `https://mcp.decisionpen.com/mcp`; S3 is needed only to access documents stored there.

## Requirements
- Remote server: `https://mcp.decisionpen.com/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http decisionpen-context-engine https://mcp.decisionpen.com/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add decisionpen-context-engine --url https://mcp.decisionpen.com/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "decisionpen-context-engine": {
            "url": "https://mcp.decisionpen.com/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "decisionpen-context-engine": {
            "type": "http",
            "url": "https://mcp.decisionpen.com/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http decisionpen-context-engine https://mcp.decisionpen.com/mcp
```

## When to use it
- locate an application or link in the architecture graph
- gather evidence and decisions attached to an entity
- find a shared standard or architecture principle
- send an ADR to the Inbox for human review

## Example prompts
- Find the `Payments` application in the architecture graph and show its links.
- Gather decisions and evidence attached to `Customer Portal`.
- Send this ADR to the Inbox for review.

## Similar
- [context-engine](https://affpapa.org/en/nevibecode/skills/context-engine.md) — skill · 585 installs
- [kibana-context-engine](https://affpapa.org/en/nevibecode/skills/kibana-context-engine.md) — skill · 422 installs
- [Code Context Engine](https://affpapa.org/en/nevibecode/mcp/code-context-engine.md) — MCP server · ★ 467 repo stars
- [Context Engine](https://affpapa.org/en/nevibecode/mcp/context-engine.md) — MCP server · ★ 324 repo stars
- [ddd-context](https://affpapa.org/en/nevibecode/skills/ddd-context.md) — skill · 739 installs
- [context-network](https://affpapa.org/en/nevibecode/skills/context-network.md) — skill · 449 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
