basemind

basemind

Give your AI coding agent a brain for your repo.

basemind is a code-map MCP server: it indexes your codebase into a queryable map so AI coding agents — Claude Code, Cursor, Continue, anything that speaks MCP — get instant semantic answers about your code. Where is this defined? Who calls it? When did it change? What's churning?

Sub-millisecond queries. 300+ languages out of the box. Local-only. Built in Rust.

---

Why your agent needs this

Today, agents read code by grepping blind. Ask Claude "who calls parseQuery?" and it ripgreps the string — you get hits in docs, tests, comments, and 14 unrelated files. The agent burns context filtering noise, then guesses.

LSPs are the semantic answer, but they're single-language, slow to start, and useless across a polyglot monorepo.

basemind is the missing layer. One index, every language, semantic-quality answers at grep speed — exposed to the agent over MCP as concrete tools (find_callers, find_references, outline, symbol_history, blame_symbol, hot_files, …) instead of "go grep again."

---

30-second setup

Install (pick one):

brew install Goldziher/tap/basemind     # macOS, Linux
npm install -g basemind                 # any Node 14+ platform
pip install basemind                    # any Python 3.8+ platform
cargo install basemind --locked         # build from source

Opt-in intelligence build (PDF/Office ingestion, semantic doc search, shared agent memory backed by LanceDB):

cargo install basemind --locked --features full

full is the meta-feature that turns on both documents (PDF / Office / HTML ingestion + OCR + layout) and memory (shared agent memory + vector search). Pulls in kreuzberg (Elastic-2.0; document parsing + bundled ONNX embeddings) and lancedb (embedded vector store). First scan after enabling downloads the embedding model into the kreuzberg cache; subsequent scans are warm.

Index your repo:

cd /path/to/your/repo
basemind scan

Wire it into Claude Code — install as a plugin:

/plugin marketplace add Goldziher/basemind
/plugin install basemind@basemind

This registers basemind as an MCP server plus a basemind skill that tells the model when to reach for code-map tools (instead of grepping or reading files one by one). Restart the session and the agent has all the tools listed below.

Codex — install via Codex's plugin / marketplace UI from the same repo (the .codex-plugin/plugin.json manifest is shipped alongside Claude's), then add the MCP server entry to ~/.codex/config.toml:

[mcp_servers.basemind]
command = "basemind"
args = ["serve"]

Other MCP clients (Cursor, Continue, Cline, …) — drop the standard mcpServers entry into the client's MCP config:

{
  "mcpServers": {
    "basemind": {
      "command": "basemind",
      "args": ["serve"]
    }
  }
}

---

What your agent gets

Code-map tools

Tool	What the agent can finally do
outline	"Give me this file's structure" — symbols, line/col, signatures, imports. One call replaces five Reads.
search_symbols	"Find anything named useAuth" — substring match across every indexed symbol, kind-filterable.
find_references	"Where is parseQuery called?" — indexed call-site lookup. No regex noise.
find_callers	"Who calls User.save()?" — resolves the definition first, then scans.
dependents	"What imports this module?" — reverse import lookup.
list_files	"What files are in src/auth/?" — indexed path + language filters.
status	"What languages does this repo use?" — file count + language breakdown.
repo_info	Branch, HEAD, workdir at a glance.

Git-aware tools