VLM Run
for LuumenAI
Pull structured data out of documents, images, and audio from the terminal
Connect VLM Run and Luumen can see the files you have uploaded, the extraction domains available, the skills your team has built, and the status of every prediction or agent run. When you need new output, it starts an extraction or an agent execution and polls until the result is ready. Luumen authenticates with an API key.
The VLM Run toolbox
11 tools: 7 read, 4 write. Reads answer instantly. Writes require approval by default. Everything is logged.
- ReadDiscover Extraction SchemasList supported structured-extraction domains, or return the full JSON schema for one domain when domain is provided.
- ReadFind FilesList uploaded files or find one by file ID or MD5 hash.
- ReadFind SkillsList VLM Run skills or find one exact skill by ID, name, and optional version.
- ReadGet RunGet the current status and result of one structured-extraction prediction or agent execution; call repeatedly to poll asynchronous work.
- ReadList AgentsReturn agents available to the connected account for selection before execution.
- ReadList ArtifactsList artifact metadata belonging to exactly one chat session or agent execution.
- ReadList RunsList structured-extraction predictions or agent executions for the connected account.
- WriteCreate SkillCreate a reusable skill from exactly one uploaded zip, prompt, or chat session. Approval by default
- WriteExecute AgentStart a VLM Run agent execution from an existing agent name or inline configuration over multimodal inputs. Approval by default
- WriteExtract Structured JSONStart structured JSON extraction from images, a document, a video, or audio using a domain, custom schema, or skill. Approval by default
- WriteUpload FileUpload a local file to VLM Run for extraction, agent input, or skill creation. Approval by default
One prompt, start to finish
What a governed VLM Run 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 skill, execute agent, extract structured json, upload file — 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 VLM Run — 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 VLM Run to work with Luumen
Connect in minutes. Every action scoped, approved, and audited from day one.