stackmap-mcp

# @ship-with-ai/stackmap-mcp

Query **[StackMap](https://stackmap.shipwithai.xyz)** — the curated map of the open-source AI stack — from any MCP client. 344 repos across 21 topics, 817 connections, every one typed (`complements` / `alternative` / `built_with`) and human-reviewed, with the *why* written down — including **when NOT to use a tool**.

No auth. No state. No telemetry. Reads the site's static JSON API; the catalog is date-versioned (`v2026.07.24` style) so you always know how fresh the map is.

## Install

**Claude Code**

```bash
claude mcp add stackmap -- npx -y @ship-with-ai/stackmap-mcp
```

**Cursor** — `.cursor/mcp.json`:

```json
{ "mcpServers": { "stackmap": { "command": "npx", "args": ["-y", "stackmap-mcp"] } } }
```

**Codex CLI** — `~/.codex/config.toml`:

```toml
[mcp_servers.stackmap]
command = "npx"
args = ["-y", "stackmap-mcp"]
```

## Tools

| Tool | What it answers |
|---|---|
| `search_catalog` | "What's on the map for *local voice transcription*?" |
| `get_repo` | "Tell me about langgraph — including when NOT to use it." |
| `get_alternatives` | "What replaces crewai, and when do I pick which?" |
| `continue_stack` | "I use langgraph + chroma — what pairs well next?" |
| `build_stack` | "Build me a starter stack for a local-first RAG pipeline over PDFs." |

Every answer carries the curator's written why — the sentence a colleague would tell you over your shoulder, not a link dump.

## How the data is made

Nothing on the map auto-publishes: an LLM proposes, a human approves, sharpens, or refuses — and both the [refusal ledger and the dismissal ledger are public](https://stackmap.shipwithai.xyz/how-we-curate). Suggest a repo at [/suggest](https://stackmap.shipwithai.xyz/suggest); dispute any edge from its repo page.

MIT · by [ShipWithAI](https://stackmap.shipwithai.xyz/about)