Skip to main content
Creation RightsCreation Rights logo icon

CR Agents Overview

Understand how AI agents work on the Creation Rights platform.

What Are CR Agents?

CR Agents are AI-powered assistants that automate tasks across the Creation Rights platform. They can process assets, clear talent, review contracts, generate reports, and perform complex multi-step operations, all without manual intervention.

Agents are designed to augment your team, not replace it. They handle the repetitive, time-consuming work so your team can focus on creative decisions.

How Agents Work

Every agent follows a simple execution model:

  1. Input. The agent receives a task with context (assets, metadata, instructions)
  2. Reasoning. The agent analyses the input and decides what actions to take
  3. Execution. The agent performs actions using platform capabilities (APIs, tools, data)
  4. Output. The agent returns results and any artifacts it produced

Agents run in a sandboxed environment with scoped permissions. They can only access the data and tools you explicitly grant them.

Agent Types

Platform Agents

Pre-built agents maintained by Creation Rights. These are production-ready and optimised for common use cases:

  • Talent Clearance Agent. Verifies talent rights and generates clearance reports
  • Asset Tagger. Analyses assets and applies metadata tags automatically
  • Contract Reviewer. Extracts key terms from contracts and flags issues
  • Report Generator. Produces usage, billing, and compliance reports

See Platform Agents for the full catalogue.

Custom Agents

Agents you build using the Agent SDK. Custom agents can:

  • Combine multiple platform capabilities
  • Integrate with external services
  • Implement domain-specific logic
  • Chain together in agent pipelines

See Building an Agent to get started.

Agent Permissions

Agents operate within a permission scope defined when they're created or installed. Scopes include:

ScopeAccess
projects:readRead project metadata and contents
projects:writeCreate and modify projects
assets:readRead asset files and metadata
assets:writeUpload and modify assets
talent:readRead talent profiles and clearance records
talent:writeCreate and update talent records
workflows:executeTrigger and manage workflows

An agent can never exceed the permissions of the user who invoked it.

Running an Agent

From the Dashboard

  1. Navigate to a project or asset
  2. Click Run Agent in the actions menu
  3. Select the agent you want to run
  4. Review and confirm the input parameters
  5. Click Run

From a Workflow

Add a Run Agent step to any workflow. The agent executes as part of the pipeline and its output is available to subsequent steps.

Via API

curl -X POST https://api.creationrights.com/v1/agents/run \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "agent_id": "talent-clearance",
    "inputs": {
      "asset_ids": ["abc123", "def456"]
    }
  }'

Monitoring Agents

View agent execution history in Agents → History. Each execution shows:

  • Status, duration, and credit cost
  • Input parameters and output results
  • Detailed execution log with reasoning steps
  • Any errors or warnings
WorkflowsBuilding an Agent