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

# ChatGrid

> Your team's project knowledge board. Drop in videos, docs, and links. AI answers from your sources. Works in your IDE.

<Note>
  **One sentence:** ChatGrid is where your team's project knowledge lives -- YouTube videos, PDFs, docs, decisions -- on a visual board that AI can search and answer from.
</Note>

## Get started in your IDE

Pick your tool. The MCP server command is the same; the config file and secret syntax are IDE-specific.

<CardGroup cols={3}>
  <Card title="Claude Code" icon="terminal" href="/mcp-setup">
    Project `.mcp.json` plus one-time approval
  </Card>

  <Card title="Cursor" icon="code" href="/mcp-setup">
    Project or global `mcp.json`
  </Card>

  <Card title="Windsurf" icon="wind" href="/mcp-setup">
    Cascade MCP config in Windsurf settings
  </Card>
</CardGroup>

```bash theme={null}
npx -y chatgrid-mcp
```

See [MCP Setup](/mcp-setup) for copy-paste configs for Claude Code, Cursor, and Windsurf.

Then talk to your AI naturally:

```
> Create a ChatGrid board for the payments project
> Add this YouTube tutorial to the board: https://youtube.com/watch?v=...
> Upload this PDF to the board: /path/to/architecture.pdf
> Search the board for webhook verification
> What does our architecture doc say about rate limiting?
```

## What can you add to a board?

Everything your team needs to remember. The API creates source nodes and vectorizes text you provide; ChatGrid MCP gives your IDE tools for creating boards, adding source nodes, uploading files, and vectorizing pasted text.

| Content type        | What happens                                                    | Example                                           |
| ------------------- | --------------------------------------------------------------- | ------------------------------------------------- |
| **YouTube videos**  | Stored as source nodes; vectorize transcripts when available    | `Add this video: https://youtube.com/watch?v=abc` |
| **Websites / URLs** | Stored as source nodes; vectorize extracted page text via API   | `Add this blog post: https://vercel.com/blog/...` |
| **PDFs**            | Uploaded to storage; process document jobs can track extraction | `Upload this PDF: /path/to/spec.pdf`              |
| **Images**          | Uploaded to storage, displayed on canvas                        | `Upload this diagram: /path/to/arch.png`          |
| **Text / Notes**    | Displayed as sticky notes, vectorized for search                | `Add a note: "We decided to use Stripe webhooks"` |

Vectorized text becomes searchable so AI can answer from your sources.

## How it works

<Steps>
  <Step title="Create a board">
    One board per project. Takes 5 seconds.
  </Step>

  <Step title="Add knowledge">
    YouTube videos, PDFs, blog posts, architecture docs, decisions. Drop them in via MCP, then vectorize the text you want AI to search.
  </Step>

  <Step title="AI answers from your sources">
    Ask questions. Get answers grounded in your actual content. Not hallucinated.
  </Step>
</Steps>

## Why ChatGrid?

Your team's knowledge dies in Slack threads and wikis nobody updates. Memory tools like claude-mem remember what **one dev** did. ChatGrid remembers what the **whole team** knows.

|                    | claude-mem | Slack             | Wiki                | **ChatGrid**                             |
| ------------------ | ---------- | ----------------- | ------------------- | ---------------------------------------- |
| Who benefits       | One dev    | Team, temporarily | Team, theoretically | **Team, permanently**                    |
| External knowledge | No         | Links that die    | Manual pages        | **Source nodes plus vectorized content** |
| AI can search it   | No         | No                | No                  | **Yes**                                  |
| Visual             | No         | No                | No                  | **Spatial canvas**                       |

## Quick start (API)

```bash theme={null}
# 1. Create a board
curl -X POST https://api.chatgrid.ai/v1/boards \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "My Project"}'

# 2. Add a YouTube source node
curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/nodes \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"type": "youtube", "position": {"x": 0, "y": 0}, "data": {"url": "https://youtube.com/watch?v=abc", "label": "My Tutorial"}}'

# 3. Add a website source node
curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/nodes \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"type": "url", "position": {"x": 400, "y": 0}, "data": {"url": "https://example.com/blog", "label": "Blog Post"}}'

# 4. Vectorize extracted text
curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/documents/vectorize \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"content": "Key notes from the tutorial and blog post...", "metadata": {"source": "quickstart"}}'

# 5. Upload a PDF
curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/assets \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@/path/to/document.pdf"

# 6. Search vectorized content
curl -X POST https://api.chatgrid.ai/v1/boards/{boardId}/documents/search \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "how does this work?"}'
```

## Next steps

<CardGroup cols={2}>
  <Card title="MCP Setup (2 min)" icon="terminal" href="/mcp-setup">
    Add ChatGrid to your IDE
  </Card>

  <Card title="Use Cases" icon="lightbulb" href="/use-cases">
    Onboarding, shared knowledge, learning
  </Card>

  <Card title="Quickstart" icon="rocket" href="/cookbooks/quickstart">
    Build your first project board
  </Card>

  <Card title="API Reference" icon="book" href="/api-reference/boards/list-boards">
    Full endpoint documentation
  </Card>
</CardGroup>