Semantic search for agents
Search the repo by meaning. Return snippets. Keep the rest out of context.
No config · any MCP client
The problem
Grep matches text, not intent. So agents guess, open file after file, and burn thousands of tokens before they find the function they actually needed. Finder indexes your repo once — embeddings plus a call graph — and answers with exact locations.
The agents
Instead of your AI client crawling the repo, it calls Finder's agents — they already know where everything lives.
Semantic search over embeddings, expanded through the call graph — BFS over callers and callees, framework edges for React, Express, Django, Spring, Gin. Field filters like kind:function or lang:rust right in the query. Optional LLM rerank.
Hands back path#Lstart-end with numbered lines and caller names — the exact snippet, nothing more. That's the token saving: the model never sees the other 99% of the file.
The kit
Finder ships agent profiles that teach your AI how to work with the index at every stage of a task — so each phase costs the fewest tokens possible.
And zero setup for your tools — one click auto-configures the MCP server and prompt files for your agent harness:
How it works
Finder reads your repo and builds a semantic map of the code.
Your agent asks a question — no need to know file names or paths.
Only the relevant lines come back. The rest of the repo never touches your context.
Languages
Run the server, point your MCP client at it, and let Finder do the reading.
Finder · semantic code search for AI agents