CocoSearch FAQs

Question 1

What makes CocoSearch privacy-first?

Accepted Answer

CocoSearch is designed for privacy. All operations, including embedding generation with Ollama and data storage with PostgreSQL, run entirely on your local machine. There are no external API calls, cloud dependencies, or telemetry involved.

Question 2

What are the main benefits of using CocoSearch for developers?

Accepted Answer

Developers can quickly onboard to new codebases, find relevant code segments using natural language, and accelerate spec-driven development. Its local-first design ensures privacy, while hybrid search and context expansion deliver highly relevant and contextual results.

Question 3

What programming languages and tools does CocoSearch support?

Accepted Answer

CocoSearch supports over 31 programming languages like Python, JavaScript, Java, C++, Go, Rust, and TypeScript. It also supports domain-specific grammars for structured config files (e.g., Dockerfile, Kubernetes YAML), leveraging Tree-sitter for syntax-aware processing.

Question 4

Can I refine my search results with CocoSearch?

Accepted Answer

Yes, CocoSearch offers powerful refinement features. You can use symbol filtering to narrow results by functions, classes, methods, or interfaces with glob patterns. It also automatically expands results to include enclosing function or class boundaries for better context.

Question 5

What is CocoSearch and how does it work?

Accepted Answer

CocoSearch is a local-first hybrid semantic code search tool. It indexes your codebases into PostgreSQL using pgvector embeddings, generated locally by Ollama. It then combines semantic similarity and keyword matching via RRF fusion to find relevant code by meaning, not just text.

CocoSearch

CocoSearch

Key Features

Use Cases

Key Features

Use Cases