# Search your documents from any assistant

Source: https://www.gryz.ai/topics/document-search/
Updated: 2026-11-03

Document search is how the AI tools you connect find the right passage in your shared context, across your documents, notes, saved memories and tasks.

**Short answer:** Connect ChatGPT, Claude or Cursor to Gryz and ask a question. Gryz search looks across your documents, notes, saved memories and tasks, by meaning and by exact words. Every result links to its source, and when nothing matches, it says so instead of guessing. There is nothing to build or host.

## Start here

- [How search works and where your data goes](https://www.gryz.ai/docs/explanation/how-search-works-and-where-your-data-goes/): How Gryz search finds your documents, notes, memories and tasks, who sees what, where your text goes, and how to turn it off.
- [MCP setup (Claude, ChatGPT, Cursor, Windsurf, Codex)](https://www.gryz.ai/docs/mcp/setup/): Step-by-step MCP connection for Claude.ai, ChatGPT, Cursor, Windsurf, and Codex with OAuth or API key authentication.
- [MCP tools reference](https://www.gryz.ai/docs/mcp/tools/): Inputs, behavior, safety annotations and limits for every Gryz MCP tool: memory, Notes, documents, Projects, Tasks, Skills, Search and company connections.

## Read more

- [Can I search my docs from ChatGPT, Claude or Cursor without building RAG?](https://www.gryz.ai/gryzle/search-docs-from-chatgpt-claude-cursor-without-rag/): Yes. No vector database, no pipeline. Gryz search finds the right passage across your docs, notes, memories and tasks from ChatGPT, Claude or Cursor.
- [Gryz vs ChatGPT memory: what's the difference?](https://www.gryz.ai/gryzle/gryz-vs-chatgpt-memory/): ChatGPT memory is great inside ChatGPT. Gryz is shared context for every agent you use: memory, docs with comments and versions, tasks, skills and search.
- [Gryz vs Notion AI: a workspace for your agents, or agents for your workspace?](https://www.gryz.ai/gryzle/gryz-vs-notion-ai/): A Notion AI alternative for your agents: Gryz gives Claude Code, Cursor and ChatGPT one shared context of memory, docs, tasks, skills and search.
- [Context compounds: an AI context layer for teams](https://www.gryz.ai/gryzle/ai-context-layer-for-teams/): Every AI tool your team uses starts from zero. A shared context layer gives people and agents one place to work from, governed by your company.
- [Chat with your documents from any AI tool](https://www.gryz.ai/gryzle/chat-with-your-documents/): Import your Word files and notes, then ask ChatGPT or Claude about them. Gryz makes your library searchable from any AI tool, privately, with no setup.
- [Shared context for AI coding agents, over MCP](https://www.gryz.ai/gryzle/shared-context-for-ai-coding-agents/): Claude Code in the morning, Cursor after lunch, Codex at night. One MCP connection gives them all the same memory, docs, tasks, skills and search.

## FAQ [#faq]

**Do I have to build RAG to search my documents?**
No. Gryz splits your documents into passages and searches them by meaning and by exact words. One MCP connection works in ChatGPT, Claude, Cursor and Codex. [This post](/gryzle/search-docs-from-chatgpt-claude-cursor-without-rag/) compares each step with building it yourself.

**What does search cover?**
Your Markdown and HTML documents, your notes, your saved memories and your tasks. You can switch notes, memories and tasks on or off. Documents are always included.

**What is never searched?**
Documents with a PIN, documents with an expiry date, documents held for review, and anything you leave out yourself.

**Can search show me something I could not already open?**
No. Results only come from what you could already open. A search in your private space never returns company content, and a company search stays in that company.

**Where does my text go?**
To prepare passages for search, Gryz sends their text to Cloudflare, and your questions go there too so they can be matched. The passages are stored in Gryz, not with Cloudflare, and Gryz does not keep the words you searched for. You can turn search off at any time. [How search works](/docs/explanation/how-search-works-and-where-your-data-goes/) has the details.

**What happens when the answer is not in my documents?**
When nothing is close enough, search says "No grounded source found". When there are only weak matches, your AI still gets the closest passages, and it is told to say it could not find the answer instead of guessing. Treat this as a guide, not a promise.