# Pumble L1 · Needs a person Pumble is a team chat tool with a documented API, command line tool and AI (MCP) server. **Can an AI agent use Pumble? No.** Our agent couldn't get into Pumble on its own. Category: Team chat · Connects by: MCP, API, CLI, SDK · Agent payments: None found · Driven: 28 Oct 2026 **How to connect** - MCP server URL: `https://mcp.pumble.com/mcp` (streamable HTTP, documented). Source: https://mcp.pumble.com/mcp - API base URL: `https://api-ga.pumble.com` (documented). Source: https://docs.pumble.com/ - CLI install: `npm install -g pumble-cli` (documented). Source: https://cake-com.github.io/pumble-node-sdk/pumble-cli.html - Docs page: `https://docs.pumble.com/` (tested in our drive). Source: https://docs.pumble.com/ **Level 1: Needs a person.** Agents can't get in on their own. - Got in by: A person sets it up (Not on its own in its last 2 drives) - Got as far as: Sign up (Use, Pay and Leave not reached) - To reach L2: Let an agent sign up on its own. 2 drives on one day, awaiting confirmation [What the levels mean](https://agentdrives.com/method/levels?via=md) ## From finding it to leaving What an agent managed without a person, at each step. 1. **Find**: Partly. Docs with gaps: llms.txt, an OpenAPI and an MCP server. 198 doc pages read. 2. **Get in**: Stopped. Can't sign up. A visible human check on sign-up; the command-line login is refused. Stopped here. 3. **Use**: Not reached. Tasks need an account. 4. **Pay**: Not reached. Buying needs an account. 5. **Leave**: Not reached. Nothing to cancel or delete. ## Getting in None of the 5 ways in worked without a person. Where it stopped: **Email and code**. A visible human check on sign-up; the command-line login is refused. - Both kinds of agent (On a server, or in its own browser): No. Signing up needs a person: it asks for a bot check. - Signed agent (Known to Cloudflare): Partly. The sign-up page opens. The check on the form still applies. | Way in | What happened | |---|---| | Email and code | Stopped by a CAPTCHA | | CLI code login | An error from the product | | Continue with Google | Not tried: Not part of this drive | | Continue with Microsoft | Not tried: Not part of this drive | | Continue with Apple | Not tried: Not part of this drive | [How agents get past bot checks](https://agentdrives.com/guides/let-signed-agents-past-your-bot-check.md?via=md) ## What an agent can do with Pumble No task was tried: our agent couldn't get into Pumble. Core tasks Pumble offers an agent: - Send a chat message (API, MCP) - Create a channel (API) - Reply in a thread (API) - React to a message (API) - Start a video call (Web app) - Search chat messages (MCP) - Share a file in chat (Web app) What it can't do, and why: - Send messages, create channels, reply, react, search or share files: Every route needs an account, and none could be made. - Buy a plan: Checkout sits inside a workspace, which needs an account. 24 more tasks it offers: - Join or leave a channel (Web app) - Change a channel's settings (Web app) - Edit or delete a message (API) - Schedule a chat message (API) - Set a status (API) - Set a reminder (Web app) - Invite a member (Web app) - Set a member's role or permissions (Web app) - Create a user group (Web app) - Kick, ban or time out a member (Web app) - Connect a third-party integration (Web app) - Import data from a file or another tool (Web app) - Build an app, bot or plugin for the product (API, CLI) - Ask the built-in AI assistant a question (Web app) - Get an AI summary or recommendation (Web app) - Build a dashboard of work data (Web app) - Give an external person guest access to a channel (Web app) - Share your screen during a call (Web app) - Record a meeting (Web app) - Create an API key (API) - Configure SAML SSO (Web app) - Read the audit log (Web app) - Work from the mobile or desktop app (Web app) - Set data retention for messages and files (Web app) **Surface audit:** 128 documented operations, 1 run, 0 passed, 1 not as documented, 127 blocked (sandbox or plan limits). ## How an agent connects MCP, API, CLI, SDK and llms.txt found. Agent surfaces: API documented · MCP documented · CLI tested · SDK documented · Docs tested · Agent payments none found. Issues found: 9. - **MCP server**: Listed, not tested. Official server. OAuth with CIMD. Sign-in needed to see its tools. https://mcp.pumble.com/mcp (7 tools across its docs) - **API**: Listed, not tested. REST Integration API, 42 operations; API addon, 26 operations. (65 operations across its docs) - **OpenAPI**: Unknown. - **CLI**: 1 tested. pumble-cli 1.1.14. (10 commands across its docs) - **SDK**: Listed, not tested. pumble-sdk 1.1.14 (Node) (39 methods across its docs) - **llms.txt**: Yes. https://pumble.com/llms.txt - **Agent payments**: None found. The checkout wasn't reached: PRO, $2.99 a month. - **Docs**: Some gaps. Covers getting started, auth and error codes. No code examples. https://docs.pumble.com/ (198 pages read) [The standards](https://agentdrives.com/standard) ## Tips for agent builders - The Integration API needs two headers: x-app-token (xpat-) and token; the API addon needs one ApiKey header, created in the app with /api-keys generate. - The MCP server at https://mcp.pumble.com/mcp uses OAuth with CIMD (PKCE S256); its metadata is at /.well-known/oauth-protected-resource/mcp. - pumble-cli 1.1.14 login fails with HTTP 400 at POST /lead; bring an existing session instead. ## How Pumble compares In Team chat: 3 of the 9 products we drove in Team chat are higher; 5 others share L1. - [Twist](https://agentdrives.com/tools/twist.md?via=md): L4, Gets in on its own - [Zulip](https://agentdrives.com/tools/zulip.md?via=md): L4, Gets in on its own - [Mattermost](https://agentdrives.com/tools/mattermost.md?via=md): L3, Gets in on its own - [Element](https://agentdrives.com/tools/element.md?via=md): L1, Can't sign up on its own - [Google Chat](https://agentdrives.com/tools/chat-google.md?via=md): L1, Can't sign up on its own - [Microsoft Teams](https://agentdrives.com/tools/microsoft.md?via=md): L1, Can't sign up on its own - **Pumble** (this page): L1, Can't sign up on its own - [Rocket.Chat](https://agentdrives.com/tools/rocket.md?via=md): L1, Can't get an API key on its own [All 9 in Team chat](https://agentdrives.com/categories/team-chat?via=md) Sorted by level, then work done, then time to the first call. [How we rank](https://agentdrives.com/method?via=md#order) On our standard (18 criteria: against other team chat products where we measured 5 or more, otherwise against every product we measured): - [A path written for agents](https://agentdrives.com/standard#docs-10): **Ahead of most** 5 of 6 others (team chat products) lower, 1 the same. - [Errors a program can read](https://agentdrives.com/standard#err-02): **Ahead of most** 84 of 104 others (every product) lower, 20 the same. - [Export before leaving](https://agentdrives.com/standard#leave-03): **In the middle** 3 of 6 others (team chat products) lower, 3 the same. - [Versions an agent can pin](https://agentdrives.com/standard#api-07): **Level with most** 59 of 97 others (every product) the same, 38 lower. - [One way to page through lists](https://agentdrives.com/standard#api-03): **In the middle** 35 of 87 others (every product) lower, 42 the same, 10 higher. - [Answers sized for an agent](https://agentdrives.com/standard#api-04): **Level with most** 61 of 85 others (every product) the same, 24 lower. - [Fast to the first call](https://agentdrives.com/standard#in-03): **Level with most** 71 of 85 others (every product) the same, 14 lower. - [Sign-in for agents acting for a person](https://agentdrives.com/standard#in-08): **The same as all** 5 others (team chat products). - [Real status codes](https://agentdrives.com/standard#err-01): **The same as all** 4 others (team chat products). - [WebMCP tools on the web app](https://agentdrives.com/standard#proto-02): **The same as all** 6 others (team chat products). - [Spend an agent can see and cap](https://agentdrives.com/standard#pay-05): **Level with most** 3 of 4 others (team chat products) the same, 1 higher. - [Pages fit one fetch](https://agentdrives.com/standard#docs-08): **Level with most** 4 of 6 others (team chat products) the same, 2 higher. - [A request id on every answer](https://agentdrives.com/standard#err-07): **Behind most** 3 of 4 others (team chat products) higher, 1 lower. - [The best way in is built for agents](https://agentdrives.com/standard#in-12): **Behind most** 4 of 7 others (team chat products) higher, 3 the same. - [Every way in works for an agent](https://agentdrives.com/standard#in-11): **Behind most** 4 of 6 others (team chat products) higher, 2 the same. - [Prices an agent can read](https://agentdrives.com/standard#find-06): **Behind most** 81 of 105 others (every product) higher, 14 the same, 10 lower. - [Errors say what to do next](https://agentdrives.com/standard#err-03): **Behind most** 80 of 103 others (every product) higher, 23 the same. - [Docs an agent can navigate](https://agentdrives.com/standard#docs-11): **Behind most** 5 of 6 others (team chat products) higher, 1 the same. [How each criterion is graded](https://agentdrives.com/standard) ## For the team behind Pumble **6 things to fix for agents.** Agent blockers: 1 found · Agent surfaces: 8 issues found. Safety review: 2 issues found. Agents get stuck at Get in. Prove Pumble is yours, then see where each issue is, read your safety findings and ask for a free drive after a fix. How to fix each one comes with the Audit. - Find: Partly - Get in: Blocks agents - Use: Not reached - Pay: Not reached - Leave: Not reached [Claim Pumble, free](https://agentdrives.com/app/products?add=pumble.com&ref=listing) · [What a drive covers](https://agentdrives.com/method?via=md) ## Where Pumble fits - Category: [Work and collaboration](https://agentdrives.com/categories/work?via=md), [Team chat](https://agentdrives.com/categories/team-chat?via=md) - Wall: [Bot checks](https://agentdrives.com/access/get-in/bot-checks) (kept our agent out), [Cloudflare bot check](https://agentdrives.com/vendors/cloudflare) (the check, and every product we met it at), [Let signed AI agents past your bot check](https://agentdrives.com/guides/let-signed-agents-past-your-bot-check?via=md) (fix guide) - Standards: [MCP](https://agentdrives.com/standards/mcp) (tools an agent can call), [llms.txt](https://agentdrives.com/standards/llms-txt) (its docs, listed for agents) - Steps: [Find](https://agentdrives.com/access/find), [Get in](https://agentdrives.com/access/get-in) (stopped here), [Use](https://agentdrives.com/access/use), [Pay](https://agentdrives.com/access/pay), [Leave](https://agentdrives.com/access/leave) - Rivals: [Element](https://agentdrives.com/tools/element?via=md) (L1), [Google Chat](https://agentdrives.com/tools/chat-google?via=md) (L1), [Microsoft Teams](https://agentdrives.com/tools/microsoft?via=md) (L1) ## Cited in 1 article - [CLI for AI agents: 11 of 14 databases have one](https://agentdrives.com/research/which-kinds-of-software-give-ai-agents-a-cli-or-mcp-server?via=md) Data: [JSON](https://agentdrives.com/api/tools/pumble.json?via=md) · [Markdown](https://agentdrives.com/tools/pumble.md?via=md) · [History](https://agentdrives.com/tools/pumble/history?via=md) · [Atom](https://agentdrives.com/feeds/tools/pumble.atom) · [JSON Feed](https://agentdrives.com/feeds/tools/pumble.json) · licence [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) **Found an error?** [Tell us on this page](https://agentdrives.com/tools/pumble?via=md#correct) or read [how corrections work](https://agentdrives.com/corrections?via=md).