Home

Columbus

The navigator your coding agent has been missing.

Columbus is a local-only, semantic code-context server your AI coding agent calls as a tool. Ask it — in plain language — where something is and get ranked, LLM-ready context with exact line ranges, reconstructed live from your working tree so it's never stale. It also owns your project's durable memory, so the agent stops re-discovering the codebase every session.

Local embeddings, no cloud, no LLM calls — natural-language search that stays on your machine.

Columbus does exactly three things:

1. Index — chart the codebase with embedded tree-sitter and embed each symbol/file on-device (metadata + git anchors + vectors, never your code).
2. Search — natural-language semantic search: vector retrieval re-ranked by deterministic heuristics, returning ranked context with exact line ranges.
3. Memory — own the project's durable record: decisions, plus structured epics → stories → tasks with history, references, and drift checks.

Why it matters

The agent is great at thinking. It's the finding and remembering that bleed tokens and go wrong. Columbus takes that off its plate.

Without Columbus	With Columbus
Greps for the exact word and misses the concept	Natural-language semantic search finds it by meaning
Reads whole files to find ten relevant lines	One call returns ranked context with exact line ranges
Context drifts; stale .md files confidently lie	Snippets rebuilt live from the working tree — never stale
Re-discovers the codebase every session	Durable memory — decisions, epics, stories & tasks persist
Repo cluttered with scattered context files	Memory is queryable and git-excluded, not committed noise
Embeddings shipped to a cloud, per-query cost	On-device embeddings, zero LLM calls, nothing leaves your machine

Start here

• New to Columbus? → Installation then Quick Start
• Want the guided tour? → Your First Index & Search
• Wiring up an agent? → Using Columbus with Your Agent
• Curious how it stays current? → Never Stale: Live Reconstruction

The one detail that explains everything

Columbus's database holds metadata, git anchors and embedding vectors only — never your code. Every snippet and exact line range is rebuilt live at query time by re-parsing the working tree; the vectors are a derived index, not your source. That single design choice is why its answers can't go stale, why it's reproducible, and why it never needs to phone home. See How Columbus Works.