# Confluence for AI agents. Managed auth, ready-made tools.

> Search, read and write Confluence pages and comments on behalf of each of your users, within the spaces they can access.

- Page: https://www.agent200.com/tools/confluence/
- Category: Files and knowledge
- Authentication: Managed OAuth
- Identifier in code: `confluence`
- Create an API key: https://agent200.com/platform/apikeys/
- Updated: 2026-11-15

## Instructions for AI agents

Connector: Confluence, through Agent200. In code: "confluence".
What it does: Search, read and write Confluence pages and comments on behalf of each of your users, within the spaces they can access.
Page: https://www.agent200.com/tools/confluence/

1. Get an Agent200 API key. If the developer has none, send them to https://agent200.com/platform/apikeys/ and ask them to choose "Create an API key". Read the key from the AGENT200_API_KEY environment variable. Never ask for it in chat, and never write it into code, prompts or logs.
2. Sign-in is managed OAuth. Each end user approves access once, on Confluence's own page, through the authorization link Agent200 returns. Never ask a user for a Confluence password or token in chat.
3. Give every request a stable id for the end user. Agent200 uses only that user's own connection.
4. Read tools: confluence_search_content, confluence_list_spaces, confluence_get_page, confluence_list_child_pages. Write tools: confluence_create_page, confluence_update_page, confluence_create_comment, confluence_add_label. Name "confluence" in the tools of a request; the model decides which tool to call.
5. Stay within the project's settings. If a tool is turned off or a call is refused, tell the user. Do not work around it.
6. Before a write, confirm with the user, unless they asked for that exact change.
7. For setup with a specific framework, follow its guide:
   https://www.agent200.com/tools/confluence/framework/claude/
   https://www.agent200.com/tools/confluence/framework/chatgpt/
   https://www.agent200.com/tools/confluence/framework/claude-code/
   https://www.agent200.com/tools/confluence/framework/muse/
   https://www.agent200.com/tools/confluence/framework/grok-bot/
   https://www.agent200.com/tools/confluence/framework/openai-dots/
   https://www.agent200.com/tools/confluence/framework/claude-duet/
   https://www.agent200.com/tools/confluence/framework/hermes-agent/
   https://www.agent200.com/tools/confluence/framework/openclaw/

The code on the connector page is a conceptual example of the SDK. SDK reference: https://docs.igpt.ai/

## About

With the Confluence connector, your agent works with each user's Confluence spaces: it searches for pages, reads them, writes new pages, updates existing ones and leaves comments. Each user connects their own Atlassian account through managed OAuth, and your agent sees only what that user can see in Confluence, within the operations your project allows.

Confluence is Atlassian's workspace for team documentation, wikis and knowledge bases.

## Confluence tools

- **Search content** (`confluence_search_content`, read): Find pages, blog posts and comments with a text or CQL query.
- **List spaces** (`confluence_list_spaces`, read): List the spaces the user can access.
- **Get a page** (`confluence_get_page`, read): Read a page's title, body and version.
- **List child pages** (`confluence_list_child_pages`, read): List the pages nested under a page.
- **Create a page** (`confluence_create_page`, write): Create a page in a space, or under a parent page.
- **Update a page** (`confluence_update_page`, write): Change a page's title or body as a new version.
- **Add a comment** (`confluence_create_comment`, write): Comment on a page or reply to a comment.
- **Add a label** (`confluence_add_label`, write): Tag a page with one or more labels.

## Limits you can set

**Turn single tools on or off.** Keep `confluence_search_content` and reading on, and switch off `confluence_update_page` if your agent should never change existing pages.

**Restrict what the agent reaches.** Choose the endpoints and scopes a project may use. A call outside them is refused.

## Get started

### Step 1: Set up the SDK

Add the Agent200 SDK to the codebase you already have and authenticate with your API key. The key covers every connector enabled for the project.

File `agent.js`:

```js
// One Agent200 API key covers every connector enabled for the project.
const agent200 = new Agent200({
  apiKey: process.env.AGENT200_API_KEY,
});
```

(Conceptual example of the Agent200 SDK.)

### Step 2: Let your user connect Confluence

Each end user authorizes their own Confluence account once. Agent200 runs the OAuth flow, handles the callback, stores the tokens and refreshes them.

File `connect.js`:

```js
// Each end user authorizes their own Confluence account, once.
const connection = await agent200
  .user("user_123")
  .connect("confluence");

if (connection.authorizationUrl) {
  // Show this link to your user; Agent200 handles the rest.
  console.log(connection.authorizationUrl);
}
```

(Conceptual example of the Agent200 SDK.)

### Step 3: Run with the Confluence tools

Name the user, the model and the tools. The model decides which Confluence tools to call, and Agent200 executes them with that user's connection.

File `run.js`:

```js
const result = await agent200
  .user("user_123")
  .run({
    model: "openai/gpt-5.6-sol",
    input: "Find our incident response runbook",
    tools: ["confluence"],
  });

console.log(result.output);
```

(Conceptual example of the Agent200 SDK.)

## Sign-in and permissions

**Who signs in.** Each of your users, once, on Atlassian's own page, choosing which Confluence site to connect. You never see or store their Atlassian password.

**What access is requested.** The scopes shown on Atlassian's consent screen, such as reading and searching content and writing pages and comments. Confluence's own space and page permissions still apply to every call.

## Example requests

- "Find our incident response runbook"
- "Summarize the requirements page for the new onboarding flow"
- "Create a meeting notes page in the Engineering space"
- "Comment on the launch plan asking who owns QA"

## Guides: Confluence with your AI framework

- [Confluence + Claude](https://www.agent200.com/tools/confluence/framework/claude.md): Connect Confluence to Claude through Agent200's MCP server: managed OAuth, ready-made tools and one API key.
- [Confluence + ChatGPT](https://www.agent200.com/tools/confluence/framework/chatgpt.md): ChatGPT cannot send an API key to a custom MCP server, so this guide builds Confluence into your own agent with OpenAI's models and the Agent200 SDK.
- [Confluence + Claude Code](https://www.agent200.com/tools/confluence/framework/claude-code.md): Have Claude Code build a Confluence agent with the Agent200 SDK: managed OAuth, ready-made tools and one API key.
- [Confluence + Muse](https://www.agent200.com/tools/confluence/framework/muse.md): Connect Confluence to Muse through Agent200's MCP server: managed OAuth, ready-made tools and one API key.
- [Confluence + Grok Bot](https://www.agent200.com/tools/confluence/framework/grok-bot.md): Connect Confluence to Grok Bot through Agent200's MCP server: managed OAuth, ready-made tools and one API key.
- [Confluence + OpenAI Dots](https://www.agent200.com/tools/confluence/framework/openai-dots.md): Dots cannot send an API key to a custom MCP server, so this guide builds an always-on Confluence agent with OpenAI's models and the Agent200 SDK.
- [Confluence + Claude Duet](https://www.agent200.com/tools/confluence/framework/claude-duet.md): Connect Confluence to Claude Duet through Agent200's MCP server: managed OAuth, ready-made tools and one API key.
- [Confluence + Hermes Agent](https://www.agent200.com/tools/confluence/framework/hermes-agent.md): Connect Confluence to Hermes Agent through Agent200's MCP server: managed OAuth, ready-made tools and one API key.
- [Confluence + OpenClaw](https://www.agent200.com/tools/confluence/framework/openclaw.md): Connect Confluence to OpenClaw through Agent200's MCP server: managed OAuth, ready-made tools and one API key.

## FAQ

### Who handles the Confluence OAuth flow and the tokens?

Agent200. It runs the authorization, handles the callback on its own domain, stores the tokens and refreshes them. You do not build or maintain an OAuth integration for Confluence.

### Does my agent act as the end user?

Within what that user authorized, yes. Every request names an end user, and Agent200 uses only that user's Confluence connection, never another user's. It cannot reach anything the user or Confluence has not authorized.

### Can I limit what my agent can do in Confluence?

Yes. You choose which services, endpoints and authorization scopes are allowed in your Agent200 configuration, and the agent works only within them.

## Related connectors

- [Google Drive](https://www.agent200.com/tools/google-drive.md): Find, read and organize the files in each user's Drive.
- [OneDrive](https://www.agent200.com/tools/onedrive.md): Search, read, upload and share OneDrive files, including Word, Excel and PowerPoint documents, on behalf of each of your users.
- [Dropbox](https://www.agent200.com/tools/dropbox.md): Search, read, upload and organize Dropbox files, and create shared links, on behalf of each of your users.
- [Notion](https://www.agent200.com/tools/notion.md): Search, read and write Notion pages and databases on behalf of each of your users.

---

Agent200: https://www.agent200.com/ · API keys: https://agent200.com/platform/apikeys/ · Docs: https://docs.igpt.ai/
