Weights & Biases Integration for AI Agents
Connect your AI agent to 35 Weights & Biases actions via MCP, A2A, or SDK, with agent authentication, tool-calling execution, and security built-in, plus 4 events to subscribe to.
Weights & Biases AI Agent Actions
35 production-ready actions for your agent to do more on Weights & Biases.
Weights & Biases Webhook Events
4 real-time Weights & Biases events your AI agent can subscribe to.
- Artifact Version Created Emitted when a new version of an artifact is logged to a project. Maps to W&B automation event `CREATE_ARTIFACT`. Fires once per `wandb.log_artifact` (or the connector's `create_artifact` action) that produces a fresh version.
- Artifact Alias Added Emitted when an alias is applied to an artifact version. Maps to W&B automation event `ADD_ARTIFACT_ALIAS`. Fires when the connector's `update_artifact` action adds one or more aliases (via the additive `addAliases` mutation), or when a user assigns an alias via the W&B SDK / UI. Filterable by regex on `alias` at the automation level.
- Artifact Tag Added Emitted when a tag is applied to an artifact version. Maps to W&B automation event `ADD_ARTIFACT_TAG`. Fires when the connector's `update_artifact` action supplies additional tag entries, or when a user tags a version via the W&B SDK / UI. Filterable by regex on `tag` at the automation level.
- Run State Change Emitted when a W&B run transitions to a new state (Running / Finished / Failed). Maps to W&B automation event `RUN_STATE`. Fires whenever the connector's `update_run` action supplies a `state` input (using the `updateRunState` mutation) and whenever the W&B SDK reports state transitions during job execution. Available on Multi-tenant and Dedicated Cloud deployments only. Payload note: W&B's RUN_STATE automation template only substitutes `run_name` (the run's display name) — no `run_id`, `run_state`, or `run_url` variable is provided. To fetch full run details, use `get_run` with (entity_name, project_name, run_name).
Do More, Build Less
Integration Infrastructure for Weights & Biases AI Agents
"What impressed us most about StackOne is its ambition and clarity. They're creating infrastructure that modern software and the entire AI agent ecosystem can rely on. The depth of secure integrations, the pace of delivery, and the team's foresight into AI's future uniquely position StackOne to redefine this category."
Luna Schmid, Partner at GV
"We've been impressed by how quickly and deeply StackOne integrates with complex enterprise systems -- and now, with their focus on agent-to-agent interoperability, they're unlocking even more powerful use cases for customers. StackOne delivers all of the above in a universal layer -- without compromise."
Barbry McGann, SVP at Workday Ventures
Weights & Biases AI agent integration resources
Webhooks for AI agents: why agentic workflows need them purpose-built
AI agents can act the moment a system changes, using webhooks. But webhooks built for software hit three limits in agentic workflows. Here is how to fix each.
7 min read
Agentic Context Engineering: Why AI Agents Kill Their Own Context Windows
AI agents exceed their context windows without knowing it. Six failure patterns and seven survival architectures for agentic context engineering.
15 min
MCP Code Mode: Keeping Tool Responses Out of Agent Context
Anthropic's code_execution processes data already in context. Custom MCP code mode keeps raw tool responses in a sandbox. 14K tokens vs 500.
11 min
Comparing BM25, TF-IDF, and Hybrid Search for MCP Tool Discovery
Benchmarking BM25, TF-IDF, and hybrid search for MCP tool discovery across 916 tools. The 80/20 TF-IDF/BM25 hybrid hits 21% Top-1 accuracy in under 1ms.
10 min
Put your AI agents to work
All the tools you need to build and scale AI agent integrations, with best-in-class connectivity, execution, and security.