Cursor/xAI personal agent (Bots on persistent cloud computer; desktop+mobile)
Official source ↗ · Source checked 2026-10-25
Check official pricing and availability. Model versions must be checked in each case.
These labels share this page. Original identifiers are preserved; versions may differ in capability or availability.
561 tag-associated cases; tool roles have not all been reviewed. This is not the product’s full capability range.
Latest case update：2026-11-02
Enter visit type, location, and time range → Grok Bot queries available slots → confirm only your chosen slot
Paste the prompt template into a Grok Bot conversation, replace the bracketed placeholders with your actual details, send, and confirm from the returned options. No extra tools or API access required.
Grok Bot
Associated by case tags. Verify costs, permissions and actual results in the case.
Paste full email → run detection prompt → get analysis of hidden links, sender mismatches, and urgent-request red flags
No additional configuration needed. Prerequisites: access to Grok Bot, and the text of the suspicious email to examine. Copy the full email (including sender headers) into the chat, then attach the detection prompt.
Grok Bot
Associated by case tags. Verify costs, permissions and actual results in the case.
Input budget and dietary restrictions → Grok Bot asks about pantry stock → generates 5-day dinner plan and store cart → human reviews and checks out
Paste the template prompt into a Grok Bot conversation, replace [budget], [allergy], and [store] with actual values, answer the pantry inventory questions, review the generated cart, and check out manually. No additional tools or API setup needed.
Grok Bot
Associated by case tags. Verify costs, permissions and actual results in the case.
Enter dietary restrictions and budget → Grok Bot generates meal plan, recipes, grocery list, and adds items to Walmart cart → user reviews and checks out
Requires a free Grok Bot account and a Walmart account. Set up via the third-party bot template "Personal Chef by Chef" on x.ai by answering six quick questions. Internal bot prompts are not visible; execution follows the template's public description.
Walmart
Associated by case tags. Verify costs, permissions and actual results in the case.
Input: natural-language request → Process: Grok Bot searches, recommends, books via ZocDoc, syncs calendar, asks questions and fills form → Output: confirmed appointment and completed intake
Enable the Doc assistant in Grok Bot and grant access to ZocDoc and Google Calendar; describe your medical need in natural language, confirm the doctor and time as prompted, then answer intake questions.
ZocDoc · Google Calendar
Associated by case tags. Verify costs, permissions and actual results in the case.
Input → paste prompt into Grok Bot chat → receive your @mail.grokbot.com address
No additional tools or setup needed. Open your Grok Bot chat, replace yourname with your desired username prefix (or keep the placeholder to see how the bot handles it), and send the prompt. Source does not specify if a particular subscription tier is required.
Grok Bot
Associated by case tags. Verify costs, permissions and actual results in the case.
Describe your transfer scenario to Grok Bot → get exact wording → copy-paste to customer service
Start a conversation in Grok Bot, describe the transfer negotiation deadlock in plain language (agent quoting higher prices than desired), and ask the bot for professional wording you can use directly. Copy the bot's output into your customer service chat. No additional setup or technical prerequisites.
Grok Bot
Associated by case tags. Verify costs, permissions and actual results in the case.
Trigger in Telegram chat → n8n routes text or voice → load Airtable memories → Grok-4 Agent calls tools → reply in Telegram
Requires an n8n instance (self-hosted or cloud), Telegram Bot token from @BotFather, xAI Grok API key, OpenAI API key, Airtable base/table, Google Calendar/Gmail OAuth, Notion integration token, and SerpAPI key. The template workflow.json can be imported directly into n8n.
n8n · telegram · grok-4 · airtable · google-calendar · gmail · notion · serpapi
Associated by case tags. Verify costs, permissions and actual results in the case.
Input: Grok Bot desktop agent with weekly quota limits → Process: Claude designs grouping / unified entry / auto-cleanup / quota gating / outcome registry → Output: 32 agents running in parallel long-term, nightly full load consuming only 7% quota
Prerequisites: Grok Bot desktop app (supports multiple persistent agents, weekly quota billing), Claude Code or CLI access, local script execution environment (codex-cli optional). Cost: Grok Bot weekly quota paid plan; Claude quota billed separately. Boundaries: Bot and CLI are separate quota pools; local Grok Bot handles one request at a time; debug-port operations rely on app-internal commands and may change with versions.
codex-cli
Associated by case tags. Verify costs, permissions and actual results in the case.
Input Bot/Routine list and budget tier → Apply a 7-step pause-inventory-prioritize-restore workflow → Output KEEP/SLOW/EVENT/CHANGE/RETIRE classification table and recovery sequence
Prerequisites: Bot list and Routine list (names + rough frequency), budget tier described as high/medium/low, Bot role classification (chat main Bot / pure radar / pure writer). Optional: third-party budget marketplace skill (read-only, with human gate recommended). Cost level: low (saves wasted quota, no extra paid skills required). Note: Source marks this as 'pending verification'; Routine names and examples are fictional. This site has not tested the workflow.
Associated by case tags. Verify costs, permissions and actual results in the case.
Input: artifact to review (game build, design doc, copy) + domain rubric → Process: experience per rubric and rank issues by impact → Output: ranked problem list with directional fixes, or a one-line pass
Prerequisites: access to grok-bot with custom bot configuration; a real artifact to review (actual game build, complete doc, or copy) and a written rubric for that domain. No additional fee conditions noted. Boundaries: Critic provides feedback only, does not edit the artifact, does not approve, and does not praise. Triggers: user sends 'Review this', a build lands, or before anything user-facing ships.
Associated by case tags. Verify costs, permissions and actual results in the case.
Use multiple Grok Bots to split content curation, drafting, review, and scheduled publishing, with shared folders for handoff and human approval as the final gate
Prerequisites: Grok Bot account with ability to create multiple bots and one-time routines; shared file system or cloud folder for bot-to-bot handoff. Inputs are scheduled briefs, shared folder paths, and scheduling rules. Known cost: medium (4 bots + one one-time routine per post + 1 nightly reconciliation routine). Underlying model version not specified.
Associated by case tags. Verify costs, permissions and actual results in the case.
Input your customer needs → Sales Lead breaks down and assigns tasks to 4 specialists → Outputs customer research, meeting briefs, follow-up drafts, and pipeline reviews, all saved to shared folder for your approval
Prerequisites: a platform account supporting multi-Bot collaboration (e.g., Grok) where you can create 5 separate Bots; access or ability to connect Gmail, Calendar, Slack, Notion, and X; ability to create a shared folder at /workspace/shared/sales/ with sub-directory permissions. Cost: not specified by source; follow the actual billing of your platform and tools. Boundaries: all outbound sending, quoting, discounting, and contract terms require user confirmation; Bots do not accept or modify meeting invites; Bots do not set price floors.
Gmail · Slack · Notion · X · Google Calendar
Associated by case tags. Verify costs, permissions and actual results in the case.
Input Google Drive presentations or documents → extract numbers per role protocol and trace earliest appearances → output provenance report with classification tags and stale alerts
Prerequisites: read access to the target Google Drive; the model must be able to access file listings and contents in Google Drive (specific connection method not documented in source, must be configured independently). Limitations: this template performs read-only analysis and never edits or updates any document; numbers that cannot be traced are marked unknown rather than deleted; numbers whose oldest traceable appearance exceeds two quarters are flagged as stale, but the operator decides whether re-measurement is needed.
google-slides · google-drive · sheets
Associated by case tags. Verify costs, permissions and actual results in the case.
Voice dictation → structured entry → impact/effort scoring → stack-ranked output
Prerequisites: access to a voice-capable grok-bot and an available GROWTH PLAYBOOK document (e.g., Notion, Google Docs, or Markdown file). This template does not own implementation, does not decide what ships, and requires no human approval. No specific integrations are listed in the source; actual connectivity depends on your platform.
Associated by case tags. Verify costs, permissions and actual results in the case.
