# Webex for AI agents. Bring the right space into focus.

> Read messages from the Webex spaces a user can access. Prepare grounded team updates and handoffs from the selected conversation.

- Page: https://www.agent200.com/connectors/webex/
- Category: Communication
- Authentication: Managed OAuth
- Identifier in code: `webex`
- Create an API key: https://agent200.com/platform/apikeys/
- Updated: 2026-11-18

## Instructions for AI agents

Connector: Webex, through Agent200. In code: "webex".
What it does: Read messages from the Webex spaces a user can access. Prepare grounded team updates and handoffs from the selected conversation.
Page: https://www.agent200.com/connectors/webex/

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 Webex's own page, through the authorization link Agent200 returns. Never ask a user for a Webex 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: webex_list_spaces, webex_get_space, webex_list_messages, webex_get_message. Name "webex" 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/connectors/webex/framework/claude/
   https://www.agent200.com/connectors/webex/framework/chatgpt/
   https://www.agent200.com/connectors/webex/framework/claude-code/
   https://www.agent200.com/connectors/webex/framework/muse/
   https://www.agent200.com/connectors/webex/framework/grok-bot/
   https://www.agent200.com/connectors/webex/framework/openai-dots/
   https://www.agent200.com/connectors/webex/framework/claude-duet/
   https://www.agent200.com/connectors/webex/framework/hermes-agent/
   https://www.agent200.com/connectors/webex/framework/openclaw/

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

## About

With the Webex connector, your agent reads messages in each user's own Webex account. It can list the spaces they belong to, read a space's details, and read the messages in a space, using only the tools your project allows.

Webex is a collaboration platform for meetings, calling and team messaging.

## Webex tools

- **List spaces** (`webex_list_spaces`, read): Find the spaces available to the connected user.
- **Read a space** (`webex_get_space`, read): Retrieve details of one identified space.
- **List messages** (`webex_list_messages`, read): Read messages in a selected space within an explicit result scope.
- **Read a message** (`webex_get_message`, read): Retrieve one identified message and its available details.

## Limits you can set

**Turn single tools on or off.** Let the agent list spaces and read messages, and switch off `webex_list_messages` if it should only see which spaces exist.

**Restrict what the agent reaches.** Choose the endpoints and scopes a project may use, for example spaces but not messages. 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 Webex

Each end user authorizes their own Webex 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 Webex account, once.
const connection = await agent200
  .user("user_123")
  .connect("webex");

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 Webex tools

Name the user, the model and the tools. The model decides which Webex 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: "List the Webex spaces I am in",
    tools: ["webex"],
  });

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

(Conceptual example of the Agent200 SDK.)

## Sign-in and permissions

**Connect the intended Webex account.** Have the end user authorize their Webex account through managed OAuth. Select a space the user belongs to before requesting messages.

**Match scopes to the read task.** Webex documents separate permissions for spaces and messages. The connected user’s membership limits the messages available to read.

**Keep user identity attached.** Pass the same authenticated application user through connection and subsequent requests. Keep the Agent200 key in server configuration.

## Example requests

- "List the Webex spaces I am in"
- "Summarize the latest messages in the Product launch space"
- "What did the team decide in the Design review space"

## Guides: Webex with your AI framework

- [Webex + Claude](https://www.agent200.com/connectors/webex/framework/claude.md): Read space messages with Claude and Agent200. Prepare a team update briefing using the connected user’s own Webex account.
- [Webex + ChatGPT](https://www.agent200.com/connectors/webex/framework/chatgpt.md): Use ChatGPT to build a separate read-only Webex assistant with Agent200. Ground the team update briefing in returned source data.
- [Webex + Claude Code](https://www.agent200.com/connectors/webex/framework/claude-code.md): Use Claude Code to build a separate read-only Webex assistant with Agent200. Ground the team update briefing in returned source data.
- [Webex + Muse](https://www.agent200.com/connectors/webex/framework/muse.md): Read space messages with Muse and Agent200. Prepare a team update briefing using the connected user’s own Webex account.
- [Webex + Grok Bot](https://www.agent200.com/connectors/webex/framework/grok-bot.md): Read space messages with Grok Bot and Agent200. Prepare a team update briefing using the connected user’s own Webex account.
- [Webex + OpenAI Dots](https://www.agent200.com/connectors/webex/framework/openai-dots.md): Use OpenAI Dots to build a separate read-only Webex assistant with Agent200. Ground the team update briefing in returned source data.
- [Webex + Claude Duet](https://www.agent200.com/connectors/webex/framework/claude-duet.md): Read space messages with Claude Duet and Agent200. Prepare a team update briefing using the connected user’s own Webex account.
- [Webex + Hermes Agent](https://www.agent200.com/connectors/webex/framework/hermes-agent.md): Read space messages with Hermes Agent and Agent200. Prepare a team update briefing using the connected user’s own Webex account.
- [Webex + OpenClaw](https://www.agent200.com/connectors/webex/framework/openclaw.md): Read space messages with OpenClaw and Agent200. Prepare a team update briefing using the connected user’s own Webex account.

## FAQ

### Which space can the agent read?

Start with a space the connected user belongs to. Identify it by its returned ID; similar space names are not proof of identity. [Webex rooms reference](https://developer.webex.com/messaging/docs/api/v1/rooms).

### Does a message list contain the whole conversation?

No. Respect pagination and any result filters. State the time window and space actually retrieved before summarizing; do not treat one page as all history.

### Can this workflow post an update?

The listed tools are read operations. A drafted update in the agent answer is not posted to Webex. Sending a message needs a separate write operation and the user’s requested destination.

### How do I handle attachments and threads?

Do not claim that file contents or every reply were read from a message summary. Request supported detail through the actual runtime when needed, and label anything not returned.

### What permissions should I check?

Confirm the connected user has access to the selected space and the read permissions for rooms and messages. Use the runtime’s advertised tool schema for actual calls; catalogue identifiers are not argument contracts.

## Related connectors

- [Slack](https://www.agent200.com/connectors/slack.md): Search conversations and post messages in the workspaces your users connect.
- [Microsoft Teams](https://www.agent200.com/connectors/microsoft-teams.md): Read channels and chats, post messages and set up meetings in Microsoft Teams on behalf of each of your users.
- [Discord](https://www.agent200.com/connectors/discord.md): Read channels, post messages and start threads in the Discord servers your users connect.
- [Zoom](https://www.agent200.com/connectors/zoom.md): Schedule, update and review Zoom meetings and their cloud recordings on behalf of each of your users.

---

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