What is a Karpathy-style LLM wiki?

It's a persistent markdown knowledge base that an AI agent can read from and write to across multiple conversations. Named after Andrej Karpathy's personal knowledge management system, it lets your LLM accumulate context and learnings over time rather than starting fresh each session.

How do I install and run this MCP server?

Install via `uvx llm-wiki-mcp@0.1.1`. It runs over stdio transport, so connect it to your LLM client (Claude, etc.) as a standard MCP server. The server will manage markdown files that persist between sessions.

Can my agent both read and write to the wiki?

Yes. The server is designed for agents to query existing wiki entries and add or update entries as they learn new information, creating a growing knowledge base.

Where are the wiki files stored?

The registry metadata doesn't specify the default storage location. Check the project repository or documentation for configuration options around where markdown files are persisted.

Is there a web UI or just markdown files?

Based on the available data, this is a markdown-based system accessed through MCP. For details on viewing or editing the wiki outside of the agent interface, refer to the project documentation.

MCP ServerSTDIOOfficialv0.1.1

Flsteven87 Llm Wiki MCP Server

Build and maintain a persistent markdown wiki that your AI agent grows over time, inspired by Andrej Karpathy's knowledge management approach. Ideal for developers who want their LLM to accumulate learnings and reference material across conversations.

io.github.flsteven87/llm-wiki-mcp

Try MCP with AI models View source

Hosted URL

Local install

Transport

STDIO

Auth

No auth required

What the Flsteven87 Llm Wiki MCP server does

How models use it and what it is built for.

Connect to Flsteven87 Llm Wiki

Local install — runs as a subprocess.

uvx llm-wiki-mcp==0.1.1

Resources

Where to find authoritative docs and source for Flsteven87 Llm Wiki.

Source repository

Example prompts for Flsteven87 Llm Wiki

Paste any of these into Agent Studio after connecting Flsteven87 Llm Wiki.

Add a new wiki entry about our API rate limits and retry strategy
Update the authentication section with the latest OAuth flow changes
Show me all wiki entries tagged with database or performance
What does the wiki say about our deployment process?

Flsteven87 Llm Wiki MCP server — FAQ

Common questions about connecting and running Flsteven87 Llm Wiki.

What is a Karpathy-style LLM wiki?
It's a persistent markdown knowledge base that an AI agent can read from and write to across multiple conversations. Named after Andrej Karpathy's personal knowledge management system, it lets your LLM accumulate context and learnings over time rather than starting fresh each session.
How do I install and run this MCP server?
Install via `uvx llm-wiki-mcp@0.1.1`. It runs over stdio transport, so connect it to your LLM client (Claude, etc.) as a standard MCP server. The server will manage markdown files that persist between sessions.
Can my agent both read and write to the wiki?
Yes. The server is designed for agents to query existing wiki entries and add or update entries as they learn new information, creating a growing knowledge base.
Where are the wiki files stored?
The registry metadata doesn't specify the default storage location. Check the project repository or documentation for configuration options around where markdown files are persisted.
Is there a web UI or just markdown files?
Based on the available data, this is a markdown-based system accessed through MCP. For details on viewing or editing the wiki outside of the agent interface, refer to the project documentation.

Skip the local setup — run MCP in your browser

MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 60+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.

Open Agent Studio