> ## Documentation Index
> Fetch the complete documentation index at: https://docs.switchagents.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> These docs moved from docs.flintai.dev to docs.switchagents.ai. Use docs.switchagents.ai for every link and request.
> To search these docs from an AI tool, connect the MCP server at https://docs.switchagents.ai/mcp. The page index is at https://docs.switchagents.ai/llms.txt.

# Use these docs

> Read the Flint AI docs from an AI tool, and find which product's pages answer a question

An AI tool can query Flint AI documentation directly — every product on this site — with no browser tab and no context switching. Ask a question where you're working and get an answer grounded in the current docs.

## Which product a question belongs to

This site documents the Flint AI product family, and the pages that answer a question depend on which product the person asking is using. Start here if that isn't settled yet.

| Product | Where it runs | Ask it about |
| - | - | - |
| [Flint AI CLI](/switch-trust-cli/index) | A terminal | Scanning an agent, evaluating one against a test run, a command that isn't behaving |
| [Flint AI Platform](/switch-trust/index) | A browser, plus an SDK and an API | Agents discovered across an organization, the models they depend on, monitoring, guardrails, cost, the Platform API |
| [Flint AI Switch](/switch-rooms/index) | A team's own messaging app | Rooms, the agents in them, shared context a whole team works from, handing work over |

Each product has its own tab in the navigation, and its landing page opens with what that product is for.

## MCP server

The Flint AI docs MCP server lets AI tools like Claude Code, Cursor, and Windsurf read documentation programmatically. Ask a question in your IDE and your agent answers from the current docs, not a stale training cutoff:

* "How do I scan a CrewAI agent with the Flint AI CLI?"
* "How do I authenticate to the Platform API?"
* "How do I remediate a hardcoded-credentials finding?"
* "How do I give a Switch room context every agent in it can work from?"

Because your coding agent can see your code as well as the docs, it can take that last question further — pulling the rule's remediation steps from the docs and applying the fix directly in your repository.

<Tabs>
  <Tab title="Claude Code">
    ```bash theme={null}
    claude mcp add --transport http switch-ai-docs https://docs.switchagents.ai/mcp
    ```
  </Tab>

  <Tab title="Cursor">
    Add to your `.cursor/mcp.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "switch-ai-docs": {
          "url": "https://docs.switchagents.ai/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="VS Code">
    Add to your VS Code settings:

    ```json theme={null}
    {
      "mcp": {
        "servers": {
          "switch-ai-docs": {
            "url": "https://docs.switchagents.ai/mcp"
          }
        }
      }
    }
    ```
  </Tab>

  <Tab title="Windsurf">
    Add to your `~/.codeium/windsurf/mcp_config.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "switch-ai-docs": {
          "serverUrl": "https://docs.switchagents.ai/mcp"
        }
      }
    }
    ```
  </Tab>
</Tabs>

Once connected, your AI tool can search across all Flint AI documentation and return grounded, cited answers in your workflow.

## Machine-readable docs

Flint AI publishes an `llms.txt` file that gives AI agents a structured index of every docs page — titles, descriptions, and URLs. AI tools use this to discover what documentation is available without crawling the site.

* [`llms.txt`](https://docs.switchagents.ai/llms.txt) — lightweight index of all pages
* [`llms-full.txt`](https://docs.switchagents.ai/llms-full.txt) — full content of all pages in a single file

## Contextual menu

Every page in the Flint AI docs includes a contextual menu that lets you send content directly to your preferred AI tool. Select the menu on any page to:

* **Copy as markdown** - paste into any AI conversation
* **View as markdown** - open the current page as raw markdown
* **Open in Claude** - send the page content to Claude
* **Open in Cursor** - load the page as context in Cursor
* **Open in VS Code** - use with Copilot or other VS Code AI tools

## Why this matters

This work happens in a terminal, an IDE, or a team's own channel — not in a browser tab. Programmatic docs access means:

* **No context switching** - ask questions without leaving your editor
* **Always current** - your AI tool reads the live docs, not cached training data
* **Specific answers** - ask a precise question like "which endpoint returns an agent's models?" and get a grounded answer, not a generic overview