Braintrust
for LuumenAI

Read Braintrust experiments and logs, then write datasets back as approved steps

Connect Braintrust and your agent can read the projects, experiments, datasets, prompts, and functions your API key can see. It can page through log events, run bounded BTQL queries, and pull aggregate summaries for an experiment. Changes — new datasets, prompt versions, inserted events, metadata edits — are previewed and approved before they land.

The Braintrust toolbox

17 tools: 5 read, 12 write. Reads answer instantly. Writes require approval by default. Everything is logged.

  • ReadFetch EventsFetch one newest-to-oldest page of project log, experiment, or dataset events.
  • ReadGet Braintrust ResourceGet one project, experiment, dataset, prompt, or function by UUID.
  • ReadList Braintrust ResourcesList and filter one page of projects, experiments, datasets, prompts, or functions visible to the connected Braintrust API key.
  • ReadQuery Braintrust DataRun one bounded synchronous read-only BTQL query over Braintrust project logs, an experiment, or a dataset.
  • ReadSummarize Experiment or DatasetReturn metadata and optional aggregate summaries for one Braintrust experiment or dataset.
  • WriteCreate DatasetCreate a named dataset under a Braintrust project, or return the existing dataset unchanged when its name already exists in that project. Approval by default
  • WriteCreate ExperimentCreate an experiment under a Braintrust project. Approval by default
  • WriteCreate ProjectCreate a Braintrust project, or return the existing project unchanged when its name already exists. Approval by default
  • WriteCreate PromptUpsert a versioned prompt configuration under a Braintrust project without invoking it. Approval by default
  • WriteDelete ProjectSoft-delete a Braintrust project and return its deletion state. Approval by default
  • WriteDelete Experiment, Dataset, or PromptDelete one Braintrust experiment, dataset, or prompt by UUID for explicit lifecycle cleanup. Approval by default
  • WriteInsert Dataset EventsUpsert a batch of examples into an existing Braintrust dataset. Approval by default
  • WriteInsert Experiment EventsUpsert a batch of result events into an existing Braintrust experiment. Approval by default
  • WriteUpdate DatasetUpdate the name, description, tags, or metadata of a Braintrust dataset. Approval by default
  • WriteUpdate ExperimentUpdate selected metadata and linkage fields on a Braintrust experiment. Approval by default
  • WriteUpdate ProjectUpdate selected fields of a Braintrust project. Approval by default
  • WriteUpdate PromptUpdate a Braintrust prompt's metadata or configuration without invoking it. Approval by default

One prompt, start to finish

What a governed Braintrust run looks like inside Luumen.

Questions

How does LuumenAI connect to Braintrust?

Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.

Can LuumenAI change things in Braintrust on its own?

Read actions answer immediately. Anything that writes — create dataset, create experiment, create project, create prompt, and more — is shown as a plan and requires approval by default, including the 2 actions classified as destructive. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.

Who gets access to the integration?

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.

Is there an audit trail?

Every call to Braintrust — 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.

Put Braintrust to work with Luumen

Connect in minutes. Every action scoped, approved, and audited from day one.