Give products and agentsreal user context they can act on
Do not rebuild assessments, parsing, memory, and access controls from scratch. Start with the Context API so OpenClaw, custom agents, and automation workflows can act on a real person instead of a prompt.
Start with context in place
Agents no longer need to reconstruct the same person on every request.
Readable, debuggable, governable contracts
Context API first, task endpoints second, with errors and field definitions stable enough to trust.
Continuous action on top of memory
OpenClaw and custom agents can filter, judge, and move work forward from the same long-term context.
Start Here
Quickstart
Get from API key to your first successful request in the shortest possible path.
Context API
Pull aggregated context first, then expand into task endpoints only when the workflow truly needs them.
Reference / OpenAPI
Use stable contracts, predictable errors, and shared field definitions to support real integration and agent debugging.
Recommended Integration PathRead context first, go deeper second
Let OpenClaw or your agent read long-term context first, then decide whether deeper task capabilities are needed.
Create an API key
Call Context API first
Expand by task
Create an API key
Generate a bearer key in settings and define access boundaries before anything else.
Call Context API first
Make the user's profile, preferences, limits, and capability signals the default context layer.
Expand by task
Add graph, alerts, prediction, and other deeper capabilities only when the workflow truly needs them.
Built for Real Workflows
AI assistants and advisors
Let agents answer from a real capability profile instead of relearning the user in every conversation.
Matching and recommendation systems
With user permission, plug ProfileClaw into matching, ranking, recommendation, and filtering workflows.
Automation workflows
Embed long-term context into notifications, coaching loops, growth plans, and recommendation systems.
OpenClaw and custom agents
Use ProfileClaw as the long-term memory layer for agents instead of scattering user identity across prompt history.
Integration Principles
Start with aggregated context before expanding into deeper task endpoints
Depend on stable types instead of free-form error strings
Keep OpenAPI and reference docs on the same contract source
Design for agent continuity, not only for single requests
Start building with real contextRead the docs, then wire it into your agents
Start with the docs, review the pricing model, and decide whether you are building for a personal agent workflow or a product integration.