grok-build-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GROK_MCP_BIN | No | Path to the grok binary | grok |
| GROK_MCP_TIMEOUT | No | Per-call timeout in milliseconds | 120000 |
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 |
|---|---|
| grok_chatA | Send a one-shot prompt to xAI Grok and return its reply. Stateless — for multi-turn use grok_consult. |
| grok_reviewA | Have Grok review a git diff. If no diff is provided, runs |
| grok_consultB | Continue a conversation with Grok by replaying the full message history each call. Stateless on the server side — the caller owns the thread. |
| grok_challengeA | Ask Grok to adversarially break a piece of code: edge cases, race conditions, security holes, adversarial inputs. Returns severity-ranked issues with reproductions. |
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 4 tools
Each tool targets a completely different interaction: adversarial challenge, one-shot conversation, multi-turn conversation, and code review. There is no overlap in purpose.
All tool names follow the identical pattern 'grok_<verb>' (challenge, chat, consult, review), providing a clear and predictable naming convention.
With four tools, the surface is lean but focused. It covers the main code interaction modes without being too few for a build-oriented MCP server.
The set covers challenging, chatting, consulting, and reviewing code. Missing a general code explanation or generation tool, but for the advertised purpose it is reasonably complete.