Bloop is a code search engine designed to enhance developer productivity by enabling natural language querying of codebases, similar to ChatGPT. It targets software engineers who need to understand, modify, or debug existing code, offering features like conversational search, code generation, and precise code navigation.

How It Works

Bloop leverages Rust for its core backend, utilizing Tantivy for search indexing and Qdrant for vector embeddings, ensuring high performance. The application is built with Tauri for cross-platform compatibility. Its semantic search capabilities are powered by on-device embeddings, prioritizing user privacy. The system supports precise code navigation for over 10 popular languages via Tree-sitter integration.

Quick Start & Requirements

The simplest way to start is by downloading the pre-built application. For building from source, cloning the repository, checking out the oss branch, and creating a local_config.json file with GitHub and OpenAI API keys is required. Git LFS must be installed for building from source.

Highlighted Details

• AI-based conversational search and code generation.
• Code Studio for LLM-powered code interaction.
• Fast regex search and symbol search.
• Precise code navigation for 10+ languages using Tree-sitter.
• On-device embeddings for privacy-focused semantic search.

Maintenance & Community

The project is actively maintained by BloopAI. Contribution guidelines and a code of conduct are available. The repository structure includes separate directories for the Tauri desktop app, the Rust backend (server/bleep), and the React frontend (client).

Licensing & Compatibility

Bloop is licensed under the Apache 2.0 license, which permits commercial use and linking with closed-source projects.

Limitations & Caveats

Building from source requires specific API keys (OpenAI, GitHub) and Git LFS. Telemetry is collected by default in the pre-built application, though it can be disabled by building from source.