#453 · Primary category: AI Coding Assistants

osgrep

colbert embeddings grep grep-search

Open Source Semantic Search for your AI Agent

Project last updated:01/17/26

GitHub Stars

1.1K

Forks

67

Contributors

14

License

Apache-2.0

Why we included this project

osgrep puts natural-language code search in the same terminal workflow where developers already use grep. Instead of matching strings, it builds a local index per repository so you can ask conceptual questions like 'where do transactions get created?' and get back the files and symbols that matter, with the role each one plays. The `trace` command is the strongest feature: it shows which functions call a target and what that target calls downstream, which is exactly the impact analysis you want before refactoring. Embeddings run entirely on your machine, so your code never leaves your laptop, and agents like Claude Code and Opencode can query the index directly instead of burning tokens reading whole files. For a team that wants its agent to answer 'where is this handled?' quickly and accurately, without sending source code to a cloud service, this is a lightweight, practical fit.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category