Your coding agent stops grepping.

Your agent reads 17 files to answer one question.

Ask it about your own docs and it greps, gets 46 matches across 17 files, then reads the biggest ones end to end. The answer was in one section. You paid for the other ten thousand tokens.

Do that sixteen times and the window is full. The agent compacts, loses the thread, and you explain what you were doing all over again.

One call. Ranked. Done.

Your agent sends one query and gets back a ranked list of the sections that answer it, not a pile of raw files to read. Underneath, Polaris indexes your markdown as both vector embeddings (via a local ONNX model) and full-text entries, runs both at query time, fuses them with RRF, and reranks with MMR for diversity. BM25 handles the lexical side; vector KNN handles semantic similarity.

Built for the way agents actually work.

Local. Private. No API key, no telemetry, no cloud round-trips, no infra to babysit. Everything runs in your process, on your machine, and stays in sync without you thinking about it. It speaks the protocol your agents already use.

One context layer. Every agent.

One .mcp.json file is all it takes. Polaris auto-configures Claude Code, Cursor, and Codex on first run.

{
  "mcpServers": {
    "polaris": {
      "command": "/path/to/polaris",
      "args": ["serve"]
    }
  }
}

Questions you're probably about to ask.

• What is Polaris? +

  A local context engine for coding agents. It indexes your project and returns, over MCP, the ranked sections that actually answer a question, about 530 tokens instead of the 10.7–15.4k a grep-and-read loop burns. The free binary indexes Markdown; Pro adds code, PDF, and .docx via polaris-ingest.

• What is MCP? +

  The Model Context Protocol, an open standard from Anthropic that lets AI agents talk to external tools. Polaris implements an MCP server so Claude Code, Cursor, and Codex can call it directly.

• What's the difference between Polaris and Polaris Pro? +

  The open-source Polaris server indexes Markdown only and is CLI-driven. Polaris Pro adds polaris-ingest for code, PDF, and .docx. A web UI for browsing the index without an agent is landing soon. Pro is available now. Start a 7-day free trial (no charge if you cancel), or buy a lifetime licence.

• Does Polaris work with Claude Code, Cursor, and Codex? +

  Yes. Any MCP-compatible client works. polaris setup writes the .mcp.json and updates CLAUDE.md / AGENTS.md / GEMINI.md automatically.

• Do I need an API key or cloud account? +

  No. Embeddings run locally via a bundled ONNX model. Nothing is sent over the network after the initial model download.

• How does Polaris compare to LangChain, LlamaIndex, or Haystack? +

  Those are full RAG frameworks for building applications. Polaris is a thin, zero-config server purpose-built for the coding-agent use case: drop it next to a repo, get MCP search.

• Can I use Polaris on private or proprietary code? +

  Yes. Everything runs locally; there's no telemetry. The MIT license permits commercial use.

• Does Polaris keep the index up to date automatically? +

  Yes. When you run polaris setup in a Claude Code project, it installs a PostToolUse hook that reindexes any changed markdown file after every Write or Edit, in about 300 ms. Your agent always searches current docs, not a stale snapshot.

• Can Polaris search on every prompt? +

  Yes. An opt-in UserPromptSubmit hook (installed with polaris setup --search-hook) searches the index on every qualifying prompt and injects the top result as context. It adds about 1 second of latency, so it is off by default.