openace
v0.1.6
Published
AI-native Contextual Code Engine — semantic code search via MCP
Maintainers
Readme
OpenACE
AI-native Contextual Code Engine. Multi-signal semantic code search via MCP for Claude Code, Codex CLI, and any MCP-compatible client.
Install
npm install -g openaceUse with Claude Code
Add to your project's .mcp.json:
{
"mcpServers": {
"openace": {
"command": "npx",
"args": ["-y", "openace", "serve", ".", "--embedding", "siliconflow"],
"env": {
"OPENACE_EMBEDDING_API_KEY": "your-api-key"
}
}
}
}Use with Codex CLI
codex mcp add openace -- npx -y openace serve . --embedding siliconflowCommands
# Start MCP server (default)
openace serve /path/to/project --embedding siliconflow
# Index a project
openace index /path/to/project --embedding siliconflow
# Search
openace search "user authentication" -p /path/to/projectMCP Tools
- semantic_search — Search code by natural language query with ranked results and snippets
- find_symbol — Look up symbols by exact name
- get_file_outline — Get structural overview of a file
How It Works
This npm package is a thin wrapper that delegates to the Python openace package via uvx. The Rust+Python engine provides:
- Multi-signal retrieval: BM25 + vector kNN + exact match + graph expansion + AST chunks
- Reciprocal Rank Fusion (RRF) for high-quality ranking
- CJK bigram tokenization for Chinese/Japanese/Korean queries
- Support for Python, TypeScript, JavaScript, Rust, Go, Java
Prerequisites
uvmust be installed (pip install uvorcurl -LsSf https://astral.sh/uv/install.sh | sh)
Custom API Providers
Supports any OpenAI-compatible embedding/reranking API:
| Variable | Description |
|----------|-------------|
| OPENACE_EMBEDDING_API_KEY | API key for embedding provider |
| OPENACE_EMBEDDING_BASE_URL | Custom base URL for embedding API |
| OPENACE_RERANKER_API_KEY | API key for reranker provider |
| OPENACE_RERANKER_BASE_URL | Custom base URL for reranker API |
License
MIT
