# jitprompt

> jitprompt is an MCP server for ai models and llm tooling. Serve an agent only the instruction sections it needs, turn by turn, from your docs. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Platforms:** Web, API
- **Website:** https://www.jitprompt.com
- **MCP Registry name:** com.jitprompt/jitprompt
- **Categories:** AI models and LLM tooling, knowledge and memory
- **Listing updated:** 2026-11-16

## About

jitprompt addresses the problem of stuffing a whole handbook into an agent's prompt. You upload documents as they are (Markdown, text, PDF or Word), and it splits them into sections with a short summary each, plus a one-line summary per document. The agent then reads a compact table of contents instead of the whole manual.

Three calls form the workflow. get_toc returns the map, get_section opens one section by ID, and search_instructions takes a description of what the agent is about to do, such as replying to a refund request for a six-week-old order, and returns the matching whole sections. Returning complete sections rather than fragments means a rule arrives with its exceptions.

The same mechanism serves skills. Upload SKILL.md files unchanged and the agent sees one line per skill, loading the full text only when a task matches. Skills are written once for any agent, whether it runs on Claude, OpenAI, Grok, LangChain or an open-weight model, and edits reach every agent on its next call. The endpoint is app.jitprompt.com/mcp; plans and prices are not published.

## Features

- **Upload as-is:** Accepts Markdown, text, PDF and Word documents.
- **Section summaries:** Each section and document gets a short summary.
- **Table of contents:** Agents read a compact map first.
- **Task-based search:** Describe the next action to get the matching sections.
- **Whole sections:** Rules arrive together with their exceptions.
- **Shared skills:** SKILL.md files load only when a task matches.

## Good for

- Give a support agent only the relevant policy section
- Share one skill library across several agent frameworks
- Keep long operating manuals out of the prompt
- Update agent guidance in one place

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http jitprompt https://app.jitprompt.com/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://app.jitprompt.com/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "jitprompt": {
      "url": "https://app.jitprompt.com/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "jitprompt": {
      "type": "http",
      "url": "https://app.jitprompt.com/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "jitprompt": {
      "serverUrl": "https://app.jitprompt.com/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "jitprompt": {
      "type": "http",
      "url": "https://app.jitprompt.com/mcp"
    }
  }
}
```

## Tools

- `get_toc`: Return the table of contents of uploaded documents
- `get_section`: Open one section by id
- `search_instructions`: Find whole sections matching a description of the planned task

## FAQ

### What file types can I upload?

Markdown, text, PDF and Word documents.

### How does an agent find the right section?

It reads the table of contents with get_toc or describes its task to search_instructions.

### Why whole sections instead of snippets?

So a rule always comes with its exceptions.

### Which agents can use it?

Any MCP-capable agent, including those on Claude, OpenAI, Grok, LangChain or open-weight models.

### How do I add jitprompt to Claude Code?

Run this command in your terminal: claude mcp add --transport http jitprompt https://app.jitprompt.com/mcp

### What are the best alternatives to jitprompt?

The closest alternatives to jitprompt are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.

## Alternatives

- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Nautilus Compass Memory](https://ai-powered.com/mcp/nautilus-compass-memory): Nautilus Compass Memory is an open-source MCP server for knowledge and memory.
- [note1](https://ai-powered.com/mcp/note1): note1 is an MCP server for productivity.
- [MedXpert 图书馆连接器（通用版）](https://ai-powered.com/mcp/medxpert): MedXpert 图书馆连接器（通用版） is an open-source MCP server for knowledge and memory.
- [Context Nest](https://ai-powered.com/mcp/context-nest): Context Nest is a freemium MCP server for knowledge and memory.
- [Tulsa Ecosystem Map](https://ai-powered.com/mcp/tulsa-ecosystem-map): Tulsa Ecosystem Map is a free MCP server for knowledge and memory.

---
Source: https://ai-powered.com/mcp/jitprompt
