Thanks to visit codestin.com
Credit goes to knitli.com

Your AI tools need
better context

CodeWeaver gives Claude, Cursor, and other AI agents precise code search—so they find what matters instead of dumping entire files into context.

Open source. MIT OR Apache-2.0.
Now in alpha.

Knitli CodeWeaver
The Beginning

CodeWeaver

Open Source MIT OR Apache-2.0 Alpha 5

A code search server that plugs into Claude, Cursor, and other AI tools via MCP (Model Context Protocol). Ask questions in plain language, get back the exact functions and classes that answer them.

What it does

  • Hybrid Search Engine: Combines keyword matching (finds exact names) with meaning-based search (finds related code). Both run automatically on every search.
  • Structural Code Parsing: Full syntax parsing using tree-sitter (27 languages) with language-aware chunking for 166+ total languages.
  • Automatic Reranking: Second-stage model rescores all results for relevance. Works with 5 different reranking providers.
  • Flexible Deployment: Choose your embedding provider: local (FastEmbed, SentenceTransformers), cloud (Voyage, AWS Bedrock, OpenAI), or bring your own. 17 providers supported.
  • Compact Tool Interface: Returns precise code snippets with metadata. Single MCP tool with ~500 tokens to describe the tool (vs. 16,000+ for LSP-based tools).

Typical Agent Workflow

Codestin Search AppVS

With CodeWeaver

By The Numbers

What Makes CodeWeaver Different

1 7
Embedding Providers
1 6 6 +
Languages Supported
~ 5 0 0
Tokens of Overhead

Always Hybrid

Every search combines keyword matching (finds exact names) with meaning-based search (finds related code). You don't configure this—it just works.

Always Reranked

Results pass through a second model that scores them for relevance. Quality defaults, no tuning required.

Broad Language Support

Full syntax parsing via tree-sitter (27 languages). Language-aware chunking for 166+ total. Practically any language that follows common patterns.

Works Offline

Run fully local with no external dependencies—or use cloud embeddings for better accuracy. Your choice.

Plugs Into Your Tools

Built on MCP (Model Context Protocol)—the standard for connecting AI tools. Works with Claude Code, Cursor, Copilot, and others.

Opinionated Defaults

Everything is configurable, but nothing needs to be configured. Sane defaults that just work.

Which Tool When?

These tools solve overlapping problems in different ways. Pick based on what you need.

MCP

Search-Only (MCP)

CodeWeaver

Plugs into your existing AI setup. Any client, Any IDE, Any workflow.

Use when:
  • You want precise control over what context your AI sees
  • You use multiple AI clients, IDEs, or workflows
  • You need fully offline/airgapped operation with good search
IDE

Full IDE

Cursor, Copilot Workspace

All-in-one: search, editing, chat, and more. Replaces your editor.

Use when:
  • You want everything in one place and don't mind switching editors
  • You're starting fresh or changing your entire workflow
IDE

IDE Extensions

Continue.dev, Cody

Adds AI or search capabilities to VS Code or JetBrains. Search + editing in your existing IDE.

Use when:
  • You want AI in VS Code and don't mind being locked to one client
  • You have an enormous repo and want something battle-tested (Cody)
MCP

LSP-Based (MCP)

Serena

Exact symbol lookup via Language Server Protocol. Search + editing via MCP.

Use when:
  • You need exact 'go to definition' style symbol lookup
  • You want IDE-like precision for supported languages
CLI

CLI Tools

Aider

Terminal-based. Repo maps for context, git integration for changes.

Use when:
  • You live in the terminal and want git-integrated AI
  • You prefer command-line workflows over GUI tools
use CodeWeaver
with all of them

The Key Tradeoffs

Codestin Search App

CodeWeaver

Hybrid search: keyword matching + meaning-based. Ask questions in plain language ('where do we handle auth?') and find relevant code even if nothing is literally named 'auth'.

Codestin Search App
Strengths
  • 166+ languages (27 with full parsing)
  • One tool, ~500 tokens overhead
  • No language server required
  • Finds conceptually related code
Codestin Search App
Tradeoffs
  • Direct symbol lookup coming soon (not yet exposed)
  • Semantic match, not always exact
Codestin Search App

Serena

Language Server Protocol (LSP)—same tech as 'go to definition' in your IDE. Search for exact symbol names (`handleAuth`) and get precisely that function.

Codestin Search App
Strengths
  • Instant, exact symbol lookup
  • IDE-like precision
  • Also does editing (9 tools)
Codestin Search App
Tradeoffs
  • ~30 languages (needs LSP for each)
  • 20+ tools, ~16,000 tokens overhead
  • Requires language server running
Codestin Search App
Bottom line: Ask questions vs. find known symbols

The full technical rundown

For those who prefer traditional comparison tables, here are the detailed specifications across all tools.

What it does
CodeWeaver: Search only
Serena: Search + edit
Cursor: Full IDE
Copilot Workspace: Full IDE
Continue.dev: IDE extension
Aider: CLI tool
How it searches
CodeWeaver: Hybrid (keyword + meaning)
Serena: Exact symbol (LSP)
Cursor: Meaning-based only
Copilot Workspace: Meaning-based only
Continue.dev: Meaning-based only
Aider: Repo maps
Reranking
CodeWeaver: Always on
Serena: No
Cursor: Varies
Copilot Workspace: Unknown
Continue.dev: Optional
Aider: No
Languages
CodeWeaver: 166+ (27 with full parsing)
Serena: ~30 (requires LSP per language)
Cursor: ~50-100
Copilot Workspace: All (text-based)
Continue.dev: ~26 (tree-sitter only)
Aider: ~165+
Context overhead
CodeWeaver: ~500 tokens
Serena: ~16,000 tokens
Cursor: N/A (built-in)
Copilot Workspace: N/A (built-in)
Continue.dev: N/A (built-in)
Aider: Varies
Offline capable
CodeWeaver: Yes (quickstart)
Serena: Yes
Cursor: Limited
Copilot Workspace: No
Continue.dev: Yes (Ollama)
Aider: Yes
Embedding providers
CodeWeaver: 17
Serena: 0
Cursor: 1-2
Copilot Workspace: 1
Continue.dev: 7
Aider: 0
Current Status

Alpha 5: Works. Not Perfect.

Solo Developer Active Development Alpha Stage

Where We Are

CodeWeaver is in Alpha 5. It works. It's not perfect. It has rough edges.

This is a solo project built by someone who believes in getting details right. Every line is written with care.

Development Timeline

Alpha Release - December 2025

Hybrid search always-on, 166+ languages, 17 providers

Alpha 5 (Current)

Stability improvements, bug fixes, structural and testing improvements

Alpha releases (Q1 2026)

Agent integration for smarter search and intent handling, data layer for including other data sources (like external API docs), documentation improvements

v1.0 (Q3 2026)

Production-ready, comprehensive docs, multi-repo

What to Expect

What Works

  • Hybrid search (keyword + meaning) across codebases
  • Automatic reranking for relevance
  • 17 embedding providers, 50+ models
  • 27 languages with code meaning
  • 166+ languages with language-aware chunking
  • Fully offline operation (quickstart profile)
  • Fallback/backup search when mainline providers fail

What's Rough

  • Documentation still improving
  • Setup requires some technical comfort
  • Performance varies by codebase size and config choices
  • Direct symbol lookup not yet exposed
  • Some edge cases not handled
CodeWeaver Knitli
The Philosophy

One Piece of the Puzzle

I built CodeWeaver because AI agents have a context problem. They re-read huge files to learn what line something was on. Most tools they're given were built for humans.

CodeWeaver is one piece of the puzzle: hybrid code search that just works. Keyword matching and meaning-based search, combined automatically. Reranking built in. Quality defaults you don't have to think about.

This is Alpha 5 software built by one person. It's rough around the edges. But the fundamentals are solid, and I'd rather ship something useful than polish forever.

Let's build it!

What's Next

  • Direct symbol lookup (LSP-style "go to definition")
  • Way better docs
  • Agent integration to assign intent and information needs, decide on search strategy
  • Performance tuning and optimization
  • Thread: dependency tracking and codebase intelligence