Top 200 ReposEXPLORER
Browse agents, categories and pages
Back to results
Planned, not doneThe source describes an intention, not a finished result.

Hand CRM updates and follow-up drafts to Grok Bot agents with approval steps

Grok Bot launched in beta in August 2026: always-on agents with their own cloud computer that sign in to your apps, learn routines by watching you, and come back when something needs your approval.

The problem

How it works

The launch post, signed by SpaceXSI on x.ai, describes Bots as teammates you message from your phone or desktop. Each Bot has its own computer in the cloud, signs in to the tools you use and works across apps and websites, including ones without an API or MCP server. A Bot can follow along while you do a job once, save it as a routine and repeat it later. Several Bots can run at once with one coordinating the rest, message each other and work together in group chats. The company’s internal examples include a sales Bot that adds call transcript notes to the CRM and drafts follow-ups, an operations Bot that processes invoices received in Gmail, and an engineering Bot that reproduces a bug, files a ticket and hands the fix to another Bot.

Results and limitations

Only launch claims and unattributed early-user quotes are available, such as one user saying they feel two to three times more efficient. There are no measured results, so this case is marked planned.

Before you try it

Bots sign in with access you give them and act in real tools, so start with draft-only or read-only permissions and keep approval on anything sent to customers. One quoted user stopped reviewing the Bot’s work; for customer-facing tasks, keep checking. If a Bot drafts outreach, follow anti-spam rules and each platform’s terms on automated activity. At launch Grok Bot was a beta for SuperGrok and eligible Cursor plans on desktop and iOS, with its own usage allowance and a waitlist for enterprises. Results depend on your setup and were not independently reproduced by Top 200 Repos.

Tools and repositories

No ranked repository is involved. This use case still stands on its own source.

Tools
Grok Bot (macOS or iOS app) · A CRM · Gmail or another email app
Prerequisites
A SuperGrok, SuperGrok Plus or SuperGrok Heavy plan, or an eligible Cursor plan, during the beta.

Try it yourself: Hand a follow-up job to an agent with clear limits

Fill the fields to adapt the prompt. Empty fields stay as visible [placeholders].

Preview
After each customer call logged in [CRM], add the key points from the call notes to the account record and draft a follow-up email in my voice. Use only what is in the notes and the account history; if something is unclear, leave a question for me instead of guessing. Do not send anything, change deal stages or contact anyone new. Put the drafts in [Where drafts go for review] and give me a list of what you changed in the CRM so I can check it. Stop and ask me if an account has a note about [Condition that needs you].

Give the Bot a separate account with limited permissions where you can, instead of your own full access.

Sources

  1. Introducing Grok Bot (opens in a new tab)

    Vendor post · x.ai · Published 11 Aug 2026 · Source checked 12 Nov 2026, 07:54 UTC

    What this source supports
    • Grok Bot launched in beta as always-on agents that have their own computer in the cloud
    • Bots sign in to existing tools, work across apps and websites without a clean API or MCP, and return when approval is needed
    • Bots can learn a workflow by following along once and save it as a routine
    • Several Bots can run in parallel, message each other and coordinate in group chats
    • Internal examples include a sales Bot updating the CRM from call notes and drafting follow-ups
    • Available to SuperGrok tiers and eligible Cursor plans on desktop and iOS, with separate usage and an enterprise waitlist
    • Includes unattributed early-user quotes, one describing a 2 to 3 times efficiency gain
Report a problem

Reported by source

Build a self-hosted AI agent workflow with human approval

Use n8n's visual canvas to wire an LLM agent to your apps, add tools and memory, and pause for human approval before it acts.

github.comn8n-io/n8n
Reported by source

Build an on-call investigation agent from your own alert history

Goutham Veeramachaneni connected Claude Code to Grafana's gcx CLI, his code and runbooks, then trained its skill on two weeks of past pages for his services.

gouthamve.dev · 21 Apr 2026anthropics/claude-code
Planned, not done

Try a desktop-operating agent safely in Anthropic's computer use demo container

Anthropic's computer-use-demo runs Claude inside a Docker desktop that you watch in the browser, with a chat panel, so you can test screen, mouse and keyboard automation away from your own machine.

github.com
Planned, not done

Delegate errands to Meta’s Muse, with approval before it sends or buys

Meta’s Muse runs in its own secure cloud VM, takes requests in WhatsApp or the Muse app, and asks before sending email or buying; a separate Sentinel agent must approve anything it sends to the internet.

about.fb.com · 8 Sept 2026
Planned, not done

Hand off research and document tasks to the Dispatch agent in Claude Cowork

Anthropic’s Cowork docs describe Dispatch, a background agent in Claude Desktop that splits one request into child tasks, runs each as its own session and forwards permission prompts to you, even from your phone.

claude.com
Reported by source

Keep a resident desktop agent for research notes and blog work

Raj Chettri runs Nous Research's Hermes Agent on his desktop for scheduled research summaries, note filing, quick scripts and his blog's writing pipeline.

rajchettri-com.pages.dev · 25 May 2026NousResearch/hermes-agent