AI Tool

Semlith

One index of everything you work on, for every agent you run.

FreemiumDeveloper Tools

ABy Aakash PawarLaunched Oct 19, 2026 UTC0 comments

Visit SemlithShare

Compare

Overview

Local search over your code and documents for AI agents.

Your agents run in a lot of places now. A terminal, an editor, a cloud session, a CI job. Each one starts from zero inside one repository, so it greps and reads whole files to find what it needs. The caller lives in another repo, the spec is in Confluence, the decision is in a Jira ticket and the diagram is in a PDF.

Semlith indexes your repositories and documents once, keeps the index fresh as files change, and answers any agent over MCP. Claude Code, Codex, Cursor and other MCP clients get back the exact file and line instead of reading whole files. It reads PDF, Word, PowerPoint and Excel too, keeps a code graph for callers and impact, and records every retrieval so you can see what each agent read.

On SWE-bench Verified, the file the fix edits is in Semlith's top 10 for 76.8% of 500 issues, against 47.0% for grep then read. Retrieval only.

Semlith runs on your machine and is free. Semlith Cloud is the paid, hosted version. It keeps one index that every agent can reach, on any laptop, in the cloud or in CI, for you alone or your whole team.

The binary is open source (Apache-2.0), has every feature and never sends anything off your machine. It runs on macOS (Apple silicon), Linux (x86_64, ARM) and Windows (x86_64), and sets itself up for 12 tested agent clients. Semlith Cloud reads GitHub, GitLab, Confluence, Jira and uploads, read only. Pro starts at $5 a month and Team at $10 a seat a month, both founding prices. Pro has a 14-day trial with no card, and Team is by invitation while Cloud opens. Queries are never metered.

Who is it for?

Developers who use AI coding agents like Claude Code, Codex and Cursor, and teams whose agents work across many repositories and documents, Startups, Founders, Developers, Indie Hackers, Small Businesses, Enterprises

Problem

AI agents only see the repository they run in. A caller in another repo, a spec in Confluence, a decision in Jira or a PDF on a drive stays out of reach, so agents grep, read whole files and guess. Nobody can see what an agent actually read.

Solution

One index of your repos and documents, kept fresh and served to every agent over MCP. Semantic and exact search, a code graph for callers and impact, readers for PDF, Word, PowerPoint and Excel, and a record of every retrieval. Semlith Cloud hosts the same index so cloud agents and CI can reach it too.

What makes it unique

The binary is free and open source with every feature, and it never sends anything off your machine. Answers come from a real code graph, not a model's guess. The benchmark is public and rerunnable. On SWE-bench Verified it puts the right file in the top 10 for 76.8% of issues, against 47.0% for grep then read.

Screenshots

Screenshot 1 of 5

Features

1One index across your repositories and documents

2Works with Claude Code, Codex, Cursor and any MCP client, 12 clients tested

3Semantic and exact (regex) search

4Code graph for callers, callees and what breaks if a symbol changes

5Impact check on GitHub pull requests with Semlith Cloud

6Reads PDF, Word, PowerPoint and Excel, so agents quote real text

7A record of every retrieval each agent made

8Re-indexes as files change

FAQ

Is Semlith free?

Yes. The binary is free and open source with every feature, and needs no account. Semlith Cloud is the paid, hosted version, from $5 a month on the founding Pro price, with a 14-day trial.

Does my code leave my machine?

No. The binary never sends anything. Semlith Cloud is opt-in and reads only the sources you connect, read only.

Which agents does it work with?

Any MCP client. Setup is tested for 12 clients, including Claude Code, Codex and Cursor. Claude, ChatGPT and Codex can also sign in to Semlith Cloud directly.

Do I need Semlith Cloud?

No, if all your agents run on your laptop. Cloud is for agents that can't reach your laptop, like cloud sessions and CI, and for sharing one index with a team.