Grok
for LuumenAI
Run Grok models from the terminal, with each generation held for approval
Connect your xAI API key and Luumen can see which models the key can reach, what they cost, and how the key is scoped. From there it drafts summaries, classifies incident text, or builds embeddings for a runbook index. Generation calls are previewed first, so you approve the model and the prompt before any tokens are spent.
The Grok toolbox
8 tools: 4 read, 4 write. Reads answer instantly. Writes require approval by default. Everything is logged.
- ReadGet API Key InfoInspect the connected xAI API key's identity, wildcard or granular ACLs, and key or team blocking state.
- ReadGet Current AccountVerify the connected xAI credential and return the caller, team, zero-data-retention status, and key blocking state.
- ReadGet VideoCheck an asynchronous xAI video request.
- ReadList ModelsList models available to the connected xAI key, including model IDs and published pricing metadata needed to choose a model for other tools.
- WriteCreate EmbeddingsCreate vector embeddings for one or more text inputs with an xAI embedding model. Approval by default
- WriteCreate ResponseGenerate a bounded, non-streaming Grok response from text or structured multimodal input, with optional reasoning, search, and function or web-search tools. Approval by default
- WriteGenerate ImageGenerate one or more images from a text prompt with an xAI image-generation model. Approval by default
- WriteGenerate VideoStart asynchronous xAI video generation from a text prompt, optionally conditioned on an image, and return a request ID for status checks. Approval by default
One prompt, start to finish
What a governed Grok run looks like inside Luumen.
Questions
Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.
Read actions answer immediately. Anything that writes — create embeddings, create response, generate image, generate video — is shown as a plan and requires approval by default. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.
You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.
Every call to Grok — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.
Often connected alongside
All integrations
Adobe Marketing Agent MCPAdobe Experience Platform context, sandbox switches, and async task status
AffindaRead parsed documents, collections, and annotations — change them with approval
Agent MailGive the agent its own mailbox for reading threads and sending updates
AI/ML APIInspect assistants, threads, and runs — and change them as approved stepsPut Grok to work with Luumen
Connect in minutes. Every action scoped, approved, and audited from day one.