Studio By Ai21 Labs MCP integration
Creates and edits AI21 assistants, validates the Python behind their plans, indexes site content, and runs an assistant on request.
57actions available
Three actions you can hand over today
Every action runs live through MCP. Nothing to build, nothing to maintain.
Run assistant
Lyro runs an AI21 Studio assistant with conversational input to handle complex customer requests. The customer gets a specialized response from a fine-tuned assistant.
Create assistant
Lyro creates a new AI assistant in AI21 Studio tailored to the team's specific use case. The assistant is ready to serve customers with domain-specific knowledge.
Ingest website connector
Lyro ingests website content via sitemap URL in AI21 Studio to build knowledge bases automatically. The assistant learns from all pages and can answer questions about the site.
How businesses use Studio By Ai21 Labs + Lyro
Each card is one request a support team gets, and the Studio By Ai21 Labs actions Lyro runs to close it.
Stand up a new AI21 assistant
Lyro creates the assistant, attaches the routes that decide where a conversation goes, and adds the plan that defines what it does, so the behavior exists before anyone opens the console.
Create AssistantCreate Assistant RouteCreate Assistant PlanKeep an assistant's website knowledge current
Lyro ingests a sitemap through a website connector, syncs it when the pages change, and reruns an ingestion that failed rather than leaving the assistant answering from a stale index.
Ingest Website ConnectorSync Website ConnectorRetry Ingest WebsiteKeep secrets and MCP storage in order
Lyro stores an API credential as a secret, updates it when the credential rotates, and registers the MCP storage an assistant reads from, so a configuration change does not wait on console access.
Create SecretUpdate SecretCreate MCP StorageRun an assistant and explain what it returned
Lyro runs an assistant with conversational input, names the resulting thread, and pulls the explanation behind an execution when someone asks why the model answered the way it did.
Run AssistantGenerate Thread NameGet Output Explanation
How it works
Get started in 3 steps
Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Lyro reads the Studio By Ai21 Labs actions it has and picks the ones a request needs.
- 01
Connect Studio By Ai21 Labs
Authorize the Studio By Ai21 Labs account your team already uses — one consent screen, no API keys, no mapping tables. Lyro can only do what you granted that account, and you can disconnect it at any time.
- 02
Tell your agent what you need
Describe the job the way you would hand it to a teammate. Lyro maps it to the Studio By Ai21 Labs actions that close it and chains as many as the request needs.
- 03
Watch it work
The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.
Get started free
Everything else about Studio By Ai21 Labs
Setup, permissions, and the limits of what Lyro can do inside Studio By Ai21 Labs.
Yes. Assistant plans in AI21 Studio carry Python, and Validate Plan checks that code on its own, so Lyro can catch a broken plan while it is being authored rather than when a live conversation hits it. Modify Assistant Plan then applies the corrected version.
Every action available in Studio By Ai21 Labs
All 57 actions your agent can call on Studio By Ai21 Labs, straight from the live MCP connection.
Check can iframe
Check if a URL can be embedded in an iframe.
Compare text
Compare two text strings and identify differences.
Create aftersales parts classification batch
Create a batch job for Fnac Aftersales parts classification.
Create assistant route
Create a new route for an AI21 Studio assistant.
Create assistant
Create a new AI assistant in AI21 Studio.
Create assistant plan
Create a new plan for an AI21 Studio assistant.
Create demo
Create a new demo in AI21 Studio.
Generate requirements
Generate requirements from a query or task description.
Generate thread name
Generate a thread name from a query or conversation context.
Create kirsh grant compliance preview
Preview grant compliance for Kirsh grants.
Kirsh grant metadata preview
Retrieve grant metadata from the Kirsh preview endpoint.
Create MCP storage
Create MCP (Model Context Protocol) storage in AI21 Studio.
Create secret
Save a secret in AI21 Studio.
Check kirsh grant compliance
Check compliance for Kirsh grant IDs.
Create website connector
Create a website connector in AI21 Studio.
Delete assistant
Delete an assistant by its unique ID.
Delete assistant route
Delete a route from an AI21 Studio assistant.
Delete demo
Delete a demo by its unique ID.
Delete MCP storage
Delete an MCP (Model Context Protocol) storage configuration by its unique ID.
Delete secret
Delete a secret by its unique ID.
Delete website connector
Permanently delete a website connector by its unique identifier.
Download modified document
Download a modified document from AI21 Labs Studio.
Get assistant
Retrieve an assistant by its unique identifier.
Get assistant route
Retrieve details of a specific route for an AI21 Studio assistant.
Get assistants by MCP
Retrieve all assistants associated with a specific MCP (Model Context Protocol) storage configuration.
Get batch prediction status
Check the status of a Fnac Aftersales parts classification batch prediction job.
Get demo
Retrieve a demo by its unique ID.
Grant kirsh metadata
Retrieve Kirsh grant metadata by grant ID.
Get library batch status
Get the ingestion status of a library batch.
Get MCP storage
Retrieve a specific MCP (Model Context Protocol) storage configuration by its unique ID.
The tools Studio By Ai21 Labs sits next to
Same connection, same setup. Pick the next one your team already uses.
Bigml
Inventories models, clusters, and anomaly detectors, reads their evaluations and predictions, and sets up projects and external data connectors.
DeepSeek
Sends prompts to deepseek-chat or deepseek-reasoner with tool calling and thinking mode, and checks balance and model availability first.
Hugging Face
Vets a model's card and security scan before adoption, runs inference, commits to Hub repositories, and watches repos through webhooks.
Modelry
Lists 3D and AR embeds, orders modeling jobs, and reports completion percentage on the ones already in progress.
Rosette text analytics
Detects the language an incoming message is written in and scores how closely two names or two addresses match before anything is acted on.
Veo
Submits a prompt as a Veo generation job, tracks the operation until it finishes, and downloads the finished clip.

Ready to connect Studio By Ai21 Labs?
Authorize the account and your agent has all 57 actions from the first conversation.


