Your agent can query this map

One command connects Claude Code, Codex, Cursor, VS Code, Gemini CLI — any MCP client — to 452 curated repos and every human-reviewed relationship, with the why written down.

Same server in every agent: npx -y @ship-with-ai/stackmap-mcp over stdio, Node.js 18+. Pick yours:

Terminal

claude mcp add --scope user stackmap -- npx -y @ship-with-ai/stackmap-mcp

--scope user makes it available in every project; without it the server registers only for the directory you ran the command in. Start a fresh session, then /mcp shows it connected.

Terminal

codex mcp add stackmap -- npx -y @ship-with-ai/stackmap-mcp

or add to ~/.codex/config.toml

[mcp_servers.stackmap]
command = "npx"
args = ["-y", "@ship-with-ai/stackmap-mcp"]

Both write the same global entry. codex mcp list shows it enabled.

Add to Cursor →

~/.cursor/mcp.json (every project) or .cursor/mcp.json (one project)

{
  "mcpServers": {
    "stackmap": {
      "command": "npx",
      "args": [
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ]
    }
  }
}

If the file already has an mcpServers object, add the stackmap entry inside it. Then check it is enabled in Cursor’s MCP settings.

Install in VS Code →

Terminal (macOS / Linux) — adds it to your user profile

code --add-mcp '{"name":"stackmap","command":"npx","args":["-y","@ship-with-ai/stackmap-mcp"]}'

or .vscode/mcp.json (this workspace)

{
  "servers": {
    "stackmap": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ]
    }
  }
}

VS Code’s file uses servers, not mcpServers. The tools show up in Copilot Chat’s agent mode.

Terminal

gemini mcp add --scope user stackmap npx -y @ship-with-ai/stackmap-mcp

Without --scope user it lands in the project’s .gemini/settings.json. Gemini connects stdio servers only in trusted folders; /mcp shows the status.

Terminal

copilot mcp add stackmap -- npx -y @ship-with-ai/stackmap-mcp

Adds it to your user configuration; copilot mcp list confirms it.

Terminal

pi mcp add stackmap -- npx -y @ship-with-ai/stackmap-mcp

Pi 1.0 has MCP built in. pi mcp list connects and lists the five tools; add -l to write the project’s .pi/mcp.json instead of ~/.pi/agent/mcp.json. Run /reload in an open session.

~/.omp/agent/mcp.json (every project) or .omp/mcp.json (one project)

{
  "$schema": "https://raw.githubusercontent.com/can1357/oh-my-pi/main/packages/coding-agent/src/config/mcp-schema.json",
  "mcpServers": {
    "stackmap": {
      "command": "npx",
      "args": [
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ]
    }
  }
}

or run the guided setup inside omp

/mcp add

Already added it to Claude Code, Codex, Gemini CLI, Cursor or OpenCode? omp reads those configs, so it is there already. /mcp reload picks up file edits; /mcp test stackmap checks the connection.

claude_desktop_config.json — macOS ~/Library/Application Support/Claude/ · Windows %APPDATA%\Claude\

{
  "mcpServers": {
    "stackmap": {
      "command": "npx",
      "args": [
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ]
    }
  }
}

Merge into the existing mcpServers object if there is one, then restart Claude Desktop. It needs Node.js installed to run npx.

~/.config/opencode/opencode.json (global) or opencode.json in the project

{
  "mcp": {
    "stackmap": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ],
      "enabled": true
    }
  }
}

OpenCode takes the command as one array. opencode mcp list shows it connected.

Zed settings.json

{
  "context_servers": {
    "stackmap": {
      "command": "npx",
      "args": [
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ]
    }
  }
}

Check Settings → AI → MCP Servers: a green dot next to stackmap means it is active.

mcp_config.json — open it from the MCPs panel in Cascade

{
  "mcpServers": {
    "stackmap": {
      "command": "npx",
      "args": [
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ]
    }
  }
}

The MCPs panel opens the right file for your install; add the entry under mcpServers and save.

cline_mcp_settings.json — open it from Cline’s MCP Servers panel

{
  "mcpServers": {
    "stackmap": {
      "command": "npx",
      "args": [
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ]
    }
  }
}

Add the entry under mcpServers and save; Cline reloads the server list.

Terminal — choose Add Extension → Command-line Extension, name it stackmap, and paste the command

goose configure
# command: npx -y @ship-with-ai/stackmap-mcp

or add to ~/.config/goose/config.yaml

extensions:
  stackmap:
    name: stackmap
    type: stdio
    cmd: npx
    args: ["-y", "@ship-with-ai/stackmap-mcp"]
    enabled: true
    envs: {}
    timeout: 300

goose calls MCP servers extensions; goose configure writes the same YAML for you.

The shape most clients accept

{
  "mcpServers": {
    "stackmap": {
      "command": "npx",
      "args": [
        "-y",
        "@ship-with-ai/stackmap-mcp"
      ]
    }
  }
}

Transport stdio, command npx, args -y @ship-with-ai/stackmap-mcp. No API key, no environment variables.

Saying “StackMap” in your question routes the agent straight to the tools.

What it answers

search_catalog — “What’s on the StackMap 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 opens with the catalog version (currently v2026.10.22) and quotes the curator’s written why — including when not to use a tool.

Nothing to trust blindly

No auth, no state, no telemetry — the server reads this site’s public JSON API and nothing else.
Public source — Ship-With-AI/stackmap-mcp, ~150 lines, CI-tested, published to npm with provenance.
Human-reviewed data — nothing on the map auto-publishes; the refusal and dismissal ledgers are public.

Prefer raw JSON? /api/catalog.json · per-repo at /api/repos/<slug>.json · llms.txt

Want this rigor applied to your team’s stack? We do paid stack audits →