# Claude Context

> Use this tool when you need to efficiently search and understand large codebases using natural language queries. It solves problems of manual code searching and discovery by providing semantic code search and indexing capabilities through vector embeddings and AST-based code splitting. With integration to multiple embedding providers and vector databases, it takes in codebases and query inputs, and outputs searchable code indexes with filtered results, making it ideal for developers working with complex codebases.

Canonical page: https://skillsregistry.net/skills/zilliz-claude-context  
JSON: https://api.skillsregistry.net/v1/skills/zilliz-claude-context

## Description

Claude Context MCP server by Zilliz provides AI assistants with semantic code search and indexing capabilities using vector embeddings and AST-based code splitting. The implementation integrates with multiple embedding providers (OpenAI, VoyageAI, Gemini, Ollama) and vector databases (Zilliz Cloud, local Milvus) to create searchable code indexes that enable natural language queries across codebases, with automatic file filtering based on extensions and ignore patterns. Built with TypeScript and supporting both VSCode extension and standalone MCP modes, it includes comprehensive evaluation framework showing 39% token usage reduction compared to grep-only approaches while maintaining equivalent retrieval quality, making it valuable for developers working with large codebases who need efficient semantic code discovery and AI-assisted code understanding workflows.

## Trust

- **Trust score (0–1):** 0.07
- **Verification tier:** scanned
- **Last scanned:** 2026-09-01

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** database
- **Updated:** 2026-09-01

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/zilliz-claude-context)
- **Repository:** <https://github.com/zilliztech/claude-context>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "zilliz-claude-context"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/zilliz-claude-context` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/zilliz-claude-context/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt