jevgrep is a command-line interface designed for coding agents to efficiently locate relevant code within large or unfamiliar repositories.
By using the Jev decision model to rank and filter files, it reduces the amount of context an agent needs to process, aiming to lower overall token costs.
- 01
Semantic code search via Jev decision model
- 02
Automated agent skill integration for Claude Code, Codex, and OpenCode
- 03
Recursive repository exploration and filtering
- 04
Local caching of search results
- 05
Support for Python, TypeScript, JavaScript, Go, and Rust declaration parsing
Open Source
The tool is free and open-source (MIT license), but users must pay their chosen AI provider for the Jev model API calls made during searches.
Best for
Large or unfamiliar codebases where coding agents would otherwise spend significant tokens on exploratory research.
Not ideal for
Small projects where agents can already find files quickly, or environments where code cannot be sent to external model providers.
A promising, specialized tool for reducing agent research costs, though its actual savings depend heavily on the specific codebase and agent behavior; users should account for both agent and Jev API costs when evaluating performance.
Version 0.5.0 was released, featuring efficiency optimizations that reduced native Jev costs by approximately 59% compared to previous versions.
As of
Is jevgrep free?+
Yes - jevgrep is Open Source. The tool is free and open-source (MIT license), but users must pay their chosen AI provider for the Jev model API calls made during searches.
Is jevgrep open source?+
Yes - jevgrep is open source. The tool is free and open-source (MIT license), but users must pay their chosen AI provider for the Jev model API calls made during searches.
Who is jevgrep best for?+
Large or unfamiliar codebases where coding agents would otherwise spend significant tokens on exploratory research.
Who is jevgrep not ideal for?+
Small projects where agents can already find files quickly, or environments where code cannot be sent to external model providers.
What are the best jevgrep alternatives?+
The closest jevgrep alternatives on ai.dosa.dev are Greptile, Fern, remio - all listed under AI Codebase Knowledge & Generation.
Who makes jevgrep?+
jevgrep is developed by David Zhang. It is listed in the AI Codebase Knowledge & Generation category on ai.dosa.dev.
Favorite this tool to revisit it later, or Zap it to contribute to the public vote count.