Integration
Connect Braintrust to Anthropic
When something happens in Braintrust, your agent does the next thing in Anthropic. Describe it in plain language and the automation is built, tested on every branch and kept running.
Build it
What do you want Braintrust and Anthropic to do together?
Describe it and your agent reads both APIs, builds the automation, tests every branch and ships it. The list below is what exists today, not the limit.
Ask the assistant
You'll sign in before anything runs. What you typed comes with you.
Or work from Claude Code
Install the APIANT plugin for Claude Code and build, edit, test and read run history from your terminal. Every plan gets this, including free, and it spends no assistant credits.
Connect Claude CodeEverything your agent can do with Braintrust and Anthropic
Triggers and actions 2 triggers ยท 26 actions
Trigger
New EventTriggers when an automation creates a new event.
Trigger
New ProjectTriggers when a new project is created.
Action
Add datasetCreate a new Braintrust dataset. POST /v1/dataset
Action
Add experimentCreate a new Braintrust experiment. POST /v1/experiment
Action
Add projectCreate a new Braintrust project. POST /v1/project
Action
Add promptCreate a new Braintrust prompt. POST /v1/prompt
Action
Delete projectDelete a Braintrust project. DELETE /v1/project/{project_id}
Action
Get datasetRetrieve a Braintrust dataset by id. GET /v1/dataset/{dataset_id}
Action
Get experimentRetrieve a Braintrust experiment by id. GET /v1/experiment/{experiment_id}
Action
Get projectRetrieve a Braintrust project by id. GET /v1/project/{project_id}
Action
Get promptRetrieve a Braintrust prompt by id. GET /v1/prompt/{prompt_id}
Action
Insert dataset eventsInsert one or more rows/events into a Braintrust dataset. POST /v1/dataset/{dataset_id}/insert
Action
Insert experiment eventsInsert one or more events (spans/rows) into a Braintrust experiment. POST /v1/experiment/{experiment_id}/insert
Action
Insert project logs eventsInsert one or more log events into a Braintrust project's logs. POST /v1/project_logs/{project_id}/insert
Action
List datasetsList Braintrust datasets. GET /v1/dataset
Action
List experimentsList Braintrust experiments. GET /v1/experiment
Action
List projectsList all Braintrust projects. GET /v1/project
Action
List promptsList Braintrust prompts. GET /v1/prompt
Action
Update projectPartially update a Braintrust project. PATCH /v1/project/{project_id}
Action
Anthropic agent messageRuns one AI agent turn on Anthropic and returns the model's reply. The turn is billed to the Anthropic account connected on this step, and the agent can call the account's AI tool automations.
Action
Cancel message batchCancel a Message Batch that is still processing.
Action
Count tokens in a messageCount the number of input tokens a Messages API request would use, without creating the message.
Action
Create message batchSubmit a batch of Messages API requests for asynchronous processing.
Action
Get message batchRetrieve the status and request counts of a Message Batch by id.
Action
Get modelRetrieve details for a specific Claude model by id.
Action
List message batchesList Message Batches in the workspace, most recently created first.
Action
List modelsList the available Claude models, most recently released first.
Action
Send model a messageSend a message to a Claude model and return its response. Supports the full Messages API: model, system prompt, multi-turn messages, max_tokens, tools, and adaptive thinking.
No trigger or action matches that. Ask your agent to add it: it reads the API documentation and builds what you describe.