@lbruton/claude-context

Fork of zilliztech/claude-context — patched for local Milvus stability.

Why this fork?

The upstream @zilliz/claude-context-mcp has recurring issues with local Milvus deployments:

• No fetch timeout on REST API calls — hangs indefinitely on DEADLINE_EXCEEDED
• No gRPC connection timeout — MilvusClient connects without a timeout, causing session startup failures
• No error differentiation — transient vs permanent failures are treated identically
• npx @latest re-downloads every session — pulls the buggy upstream on each new Claude Code session

This fork fixes those issues and publishes to @lbruton/claude-context-mcp on npm so sessions always use the patched version.

Changes from upstream

1. 30s fetch timeout on all REST API requests (AbortSignal.timeout)
2. 30s gRPC connection timeout on MilvusClient initialization
3. Better error logging — timeout vs error differentiated in log output
4. Rebranded to @lbruton/ npm scope — @lbruton/claude-context-core + @lbruton/claude-context-mcp

Installation (Claude Code MCP)

In ~/.claude/.mcp.json:

{
    "mcpServers": {
        "claude-context": {
            "command": "npx",
            "args": ["-y", "@lbruton/claude-context-mcp@latest"],
            "env": {
                "OPENAI_API_KEY": "sk-...",
                "EMBEDDING_PROVIDER": "OpenAI",
                "EMBEDDING_MODEL": "text-embedding-3-small",
                "MILVUS_ADDRESS": "192.168.1.81:19530"
            },
            "startupTimeoutMs": 30000
        }
    }
}

Environment Variables

| Variable | Required | Description | | -------------------- | -------- | -------------------------------------------------- | | MILVUS_ADDRESS | Yes | Milvus gRPC endpoint (e.g., localhost:19530) | | OPENAI_API_KEY | Yes* | OpenAI API key for embeddings | | EMBEDDING_PROVIDER | No | OpenAI (default), VoyageAI, Gemini, Ollama | | EMBEDDING_MODEL | No | Model name (default: text-embedding-3-small) | | MILVUS_TOKEN | No | Milvus authentication token (optional) |

*Required when using OpenAI embeddings (default).

Local Milvus Setup

This fork is designed for local Milvus Standalone. Run Milvus via Docker:

# Milvus Standalone (includes etcd + minio)
docker compose -f docker-compose-milvus.yml up -d

Or use an existing Milvus instance on your network — just set MILVUS_ADDRESS to point to it.

Building from source

pnpm install
pnpm build:core
pnpm build:mcp

Publishing

pnpm release:core
pnpm release:mcp

License

MIT — same as upstream.

Upstream

Original project: zilliztech/claude-context