LLM & AI Access

How to access Orca documentation for AI tools, LLMs, and programmatic use

LLM & AI Access

Orca documentation is structured for direct consumption by LLMs and AI tools. Whether you’re grounding a language model with protocol context or building an autonomous agent, the resources below give you everything you need.

MCP Server

Orca exposes a live Model Context Protocol (MCP) server — the fastest way to give any AI tool native access to Orca documentation.

https://docs.orca.so/mcp

The MCP server provides a SearchOrcaDocumentation tool that does semantic search across all docs and returns titles, content excerpts, and direct page links. No API key required.

Connect in Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "orca-docs": {
      "url": "https://docs.orca.so/mcp"
    }
  }
}

Connect in Cursor

Add to your .cursor/mcp.json:

{
  "mcpServers": {
    "orca-docs": {
      "url": "https://docs.orca.so/mcp"
    }
  }
}

Connect in any MCP-compatible client

The server uses the standard MCP HTTP transport. Initialize with:

curl -s -X POST https://docs.orca.so/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"my-agent","version":"1.0"}},"id":1}'

Then call the search tool:

curl -s -X POST https://docs.orca.so/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"SearchOrcaDocumentation","arguments":{"query":"how to open a CLMM position"}},"id":2}'

---

Quick Access Files

File	URL	Description
llms.txt	docs.orca.so/llms.txt	Full structured reference: protocol constants, key concepts, SDK function signatures, and page links — optimized for LLM context

Using with AI Tools

ChatGPT, Claude, and Other LLMs

Paste this prompt to load Orca context into any LLM:

Fetch https://docs.orca.so/llms.txt and use it to answer questions about Orca.

llms.txt includes protocol constants, fee tiers, SDK function signatures, price math formulas, and links to all documentation pages — everything an LLM needs to reason about Orca accurately.

Cursor, Windsurf, and AI Code Editors

Add the MCP server (above) for native tool-use, or reference llms.txt directly:

1. Cursor: Add https://docs.orca.so/llms.txt as a doc source in Settings → Docs
2. Windsurf: Include the URL in your workspace context
3. VS Code Copilot: Reference the URL directly in your prompts

Programmatic Access

# Fetch the full structured reference
curl https://docs.orca.so/llms.txt

llms.txt Standard

Orca follows the llms.txt specification, an emerging open standard that makes documentation natively accessible to AI systems — analogous to robots.txt for search crawlers. The format provides:

• Protocol constants — program IDs, config addresses, fee tiers, error codes
• Key concepts — tick math, sqrtPrice formulas, position lifecycle
• SDK function signatures — ready-to-use TypeScript Kit references
• Organized page links — every documentation page with a one-line description

Best Practices

When using Orca docs with AI tools:

1. Ground with llms.txt first — load it as context before asking protocol-specific questions
2. Reference specific pages — for focused queries, link directly to the relevant documentation page
3. Use the REST API for live data — llms.txt contains static reference; fetch https://api.orca.so/v2/solana/pools/search?q=SOL-USDC for real-time pool state

The documentation is continuously updated. AI tools accessing these URLs will always receive the latest content.

---

Building an autonomous agent on Orca? See AI Agents on Orca for a complete guide.