Agents API

Use this section for twins, definitions, workspaces, prompt lab, and analytics. Automation moved to Chain Triggers — a Chain is the executable workflow and a trigger declares when it runs.

Base Path

https://api.chainabit.com/api/v1

Agent resources are split across two base paths:

• /agents/* for global agent management
• /workspaces/:workspaceId/ai/* for workspace-scoped AI resources

Authentication

All Agents endpoints require a valid JWT Bearer token and an active entitlement.

Environment Variables

export BASE_URL="https://api.chainabit.com/api/v1"
export TOKEN="your-access-token"
export WORKSPACE_ID="your-workspace-id"

Sections

• Conversation Participants & Lenses for inviting agents into a conversation, @mention routing, importing LenserFight agents, and calling Lenses in a message.
• Twins for twins, memories, and twin actions.
• Twin Brain for the Brain Simulator — brain graph, memory graph, live activity, and narration.
• Definitions for definitions, versions, instances, scopes, and tools.
• Workspaces for workspace agents, tools, skills, and twin policies.
• Prompt Lab for prompt templates and tests.
• Analytics for workspace analytics.
• Contexts for managing curated knowledge injected into twin prompts.

Notes

• Workspace-scoped endpoints require membership in that workspace.
• Definitions are the starting point for reusable agents.
• Skills must be published before they can be used by agents.