searchLABSCO SUMMARY
Two tools, and the reason to use them is what you stop sending to the model.
The claim the project leads with is token cost: returning matched chunks instead of grep hits plus whole files. Indexing an average repo takes about 500 ms and queries about 1 ms on CPU, with NDCG@10 of 0.854 on their own benchmarks — self-reported, but measurable on your own repo since there is nothing to sign up for.
What it is
A code search library built for agents, exposed as an MCP server. Ask it a question in plain language and it returns the matching chunks rather than whole files, which is what makes it cheap enough for an agent to use constantly. Everything runs locally on CPU: no API keys, no GPU, no external service.
What you get
- A codebase searched with a natural-language or code query, against a local path or an
https://git URL, scoped tocode,docs,configorall—search - Chunks semantically similar to a given file and line, for finding related implementations —
find_related - On-demand indexing with caching, and automatic re-indexing of local paths when files change
- A standalone CLI for scripts and non-MCP use, including searching a remote repo that is cloned on demand, and a
--max-snippet-linescontrol for how much of each hit comes back - Index control through
.gitignoreand.sembleignore, where a!prefix force-includes extensions it would otherwise skip - An interactive installer that detects your agents and offers three integrations: the MCP server, CLI guidance written into AGENTS.md or CLAUDE.md, and a dedicated
semble-searchsub-agent
Requirements
uv, then uv tool install semble and semble install to wire it into your agents; semble uninstall reverses it. No API keys, no GPU, no external services — everything runs on CPU. Indexes are built and cached on first run and invalidated automatically when files change. MIT licensed.
Setup effort
One command — uv tool install semble && semble install
USE CASES
01
Asking how something works in a codebase you do not know
Query it the way you would ask a colleague — "how is authentication handled" — and get the relevant functions back, not a list of files to open.
02
Finding the other places that do the same thing
Point at a file and line and get semantically similar chunks, which is how you catch the three near-duplicate implementations before adding a fourth.
03
Looking something up in a repo you have not cloned
Pass a git URL instead of a path; it clones and indexes on demand and caches the result.
04
Searching docs or config rather than code
Set the content scope to docs, config or all — the deployment guide is not in the source tree.
2 tools
Listed in project docs. Your client may show a different list.
find_relatedBEFORE YOU INSTALL
What to know first
Tools you get
2 tools named in the project's documentation.
Requires
uvx on PATH
Where it runs
All 4 documented clients run the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
uv tool install semble
semble installMCP Clients
Claude Code, Cursor, Other MCP clients
JSON
{
"mcpServers": {
"semble": {
"command": "uvx",
"args": [
"--from",
"semble[mcp]",
"semble"
]
}
}
}VS Code
VS CODE · JSON
{
"servers": {
"semble": {
"command": "uvx",
"args": [
"--from",
"semble[mcp]",
"semble"
]
}
}
}
