grok-bot-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| check_usageA | Check weekly usage, quota percentage, remaining requests, and on-demand spend for Cursor / Grok Bot accounts. If account='all', returns usage for all saved accounts. |
| switch_accountA | Switch the active account for Cursor / Grok Bot across both CLI/MCP and the macOS Grok Bot desktop application. Next time you open the Grok Bot desktop app, it will launch directly into this account. |
| list_botsA | List all Grok Bot agents (bots) currently running or registered on this machine. Returns agent IDs, names, origins, and descriptions. |
| create_botC | Create a new Grok Bot agent instance with a specific name and task description. |
| delete_botC | Permanently delete a Grok Bot agent by its agent ID. |
| send_messageB | Send a message or user prompt to a Grok Bot agent and receive its response. |
| get_transcriptB | Read the conversation transcript of a Grok Bot agent in chronological order. |
| search_botsB | Search for Grok Bot agents matching a keyword in their name or description. |
| list_databasesA | List all local Grok Bot SQLite database files on this machine, showing agent IDs and file sizes. |
| read_transcript_entriesB | Directly read raw transcript entries from a bot's local SQLite database without network overhead. |
| search_messagesC | Search across all local bot transcripts for messages containing a keyword. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 11 tools
Most tools target clear distinct operations (account vs bot lifecycle vs transcript vs database). However, get_transcript, read_transcript_entries, and search_messages all operate over transcript data, and list_bots vs search_bots overlap, which could cause occasional misselection.
All 11 tools follow a consistent snake_case verb_noun pattern (check_usage, switch_account, list_bots, create_bot, delete_bot, send_message, get_transcript, search_bots, list_databases, read_transcript_entries, search_messages). No mixed conventions.
11 tools is well within the ideal 3-15 range and each covers a meaningful slice of the domain (accounts, bots, transcripts, databases). Nothing feels redundant or padded.
Core lifecycle is covered: list/search/create/delete bots, send/get messages, and query databases. Minor gaps exist, e.g. no update_bot and no tool to add/list accounts beyond check_usage/switch_account, but agents can work around these.