Gemini
for LuumenAI
Text, embeddings, images, and Veo video from one Gemini connection
Connect Gemini and your agent can draft text, count tokens before a long run, generate embeddings, and check which models are available with their limits. It can also produce images with Nano Banana and videos with Veo, then poll the job and hand back the finished file. Generation steps are previewed and approved before they run.
The Gemini toolbox
7 tools: 5 read, 2 write. Reads answer instantly. Writes require approval by default. Everything is logged.
- ReadCount Tokens (Gemini)Counts the number of tokens in text using Gemini tokenization.
- ReadEmbed Content (Gemini)Generates text embeddings using Gemini embedding models.
- ReadGenerate Content (Gemini)Generates text content or speech audio from prompts using Gemini models.
- ReadList Models (Gemini API)Lists available Gemini and Veo models with their capabilities and limits.
- ReadWait and Download Video (Veo)Polls a Veo video generation operation until completion, then downloads and returns the video file.
- WriteGenerate Image (Nano Banana)Generates images from text prompts using Gemini models (Nano Banana). Approval by default
- WriteGenerate Videos (Veo)Generates videos from text prompts using Google's Veo models. Approval by default
One prompt, start to finish
What a governed Gemini run looks like inside Luumen.
Questions
Gemini needs no credential of its own, so there is nothing to authorize. Access is still granted per agent, skill, and team inside Luumen, and every call is logged.
This connection exposes write actions such as generate image (nano banana), generate videos (veo). For connected tools, admins configure which actions may run automatically and which require approval, within the access granted to the user. Review the configured approval policy before enabling write access.
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 Gemini — 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 Gemini to work with Luumen
Connect in minutes. Every action scoped, approved, and audited from day one.