API
Last updated
Last updated
Here’s a full breakdown of how to configure and set up the API Skill in raiaAI’s Launch Pad. This skill allows you to interact with your AI agent programmatically via RESTful API endpoints—perfect for integrating the agent into custom software, workflows, or apps.
The API Documentation can be .
Each Agent has an API Key which you generated under the API Skill when editing the Agent,
Pass the API Key via the header as Agent-Secret-Key
Example
The API Skill gives developers access to:
Agent-level endpoints (create/update/manage agents)
Conversation-level endpoints (start, fetch, and update threads)
Prompt & response data (enable fine-tuned logic or storage)
You can:
Send user messages to your agent and get responses
Retrieve transcripts or logs
Create custom interfaces or workflows around the agent
1. Enable the API Skill
Go to Agent → Skills
Click Add Skill
Select API Skill and toggle it Active
2. Generate API Credentials
Once the skill is enabled:
Navigate to Settings (or the "API & Integrations" tab if present)
Click “Generate Secret Key”
This is your Bearer Token for authorization
Store this securely — it's required for all requests
You may also receive:
Agent ID
Organization ID (for multi-agent setups)
3. Available Endpoints (Typical Examples)
Here’s what you can do:
POST /api/conversations/start
Start a new thread
POST /api/conversations/:id/message
Send message to an existing conversation
GET /api/conversations/:id
Get conversation data
GET /api/agents/:id
Fetch agent info
PUT /api/agents/:id
Update agent configuration
GET /api/prompts/:id
Retrieve prompt data
GET /api/threads
List all conversations (admin only)
4. Authentication
All requests must include an Authorization header:
You can also assign keys per organization or per agent depending on access controls.
5. Webhook Integration (Optional)
For two-way communication, combine with the Webhook Skill to:
Receive messages or summaries asynchronously
Trigger real-time updates to external apps after agent replies
6. Test API Usage
Use tools like:
Postman or cURL for quick testing
Swagger UI (if available) for interactive exploration
Verify:
Agent responds as expected
Secure access is enforced
Rate limits or usage quotas (if applicable)
Rotate keys regularly
Enhances security
Use unique keys per integration
Isolates systems and simplifies auditing
Limit API scope
Follow least-privilege principle for endpoints
Combine with Scoring & Webhooks
Enables powerful analytics & follow-up workflows