Zread Mcp
Let Claude Code read and search upstream GitHub repo documentation on zread.ai before you fork or integrate a library.
Overview
Zread MCP is an MCP server for the Idea phase that lets AI assistants read zread.ai-hosted GitHub docs, search code, and chat with repository AI.
What is this MCP server?
- Stdio MCP server (npm zread-mcp v1.0.1) with ZREAD_TOKEN secret env
- Read GitHub repository documentation hosted on the zread.ai platform
- Search code across indexed repos from inside the agent
- Conversational Q&A with per-repository AI on zread.ai
- Bridges agent context to external doc index without cloning every repo locally
- Server version 1.0.1
- 1 npm stdio package (zread-mcp)
- 1 required secret env var (ZREAD_TOKEN)
Community signal: 2 GitHub stars.
What problem does it solve?
Agents only see what is in your workspace, so evaluating unfamiliar GitHub repos means endless tab-switching and incomplete context.
Who is it for?
Solo builders comparing OSS libraries or onboarding to a large upstream repo before writing integration code.
Skip if: Teams that only need docs from repos already cloned locally with full-text search in the IDE.
What do I get? / Deliverables
Your agent can query zread.ai for repo docs and code search in-session so research moves faster toward a build decision.
- In-agent access to zread-indexed repo documentation
- Code search and repo AI dialogue from the agent session
Recommended MCP Servers
Journey fit
Solo builders research open-source repos and third-party SDKs in the Idea phase before committing to a stack. Research subphase is where you dig into competitor codebases, README depth, and API surface—not yet building.
How it compares
MCP doc-and-search bridge to zread.ai, not a local ripgrep or GitHub clone workflow.
Common Questions / FAQ
Who is zread-mcp for?
Indie developers and agent users who research GitHub projects through zread.ai and want that corpus inside Claude Code, Cursor, or similar MCP clients.
When should I use zread-mcp?
Use it during idea and research work when you need indexed documentation, code search, or repo AI chat without leaving your agent session.
How do I add zread-mcp to my agent?
Add the npm package zread-mcp (stdio), set the ZREAD_TOKEN environment variable to your zread.ai API key, and register the server in your client’s MCP config.