Automation Tools

You can integrate Cencori into your automation workflows using standard HTTP requests. This allows you to add secure, governed AI to any business process.

n8n is a powerful open-source workflow automation tool. You can use the HTTP Request node to connect to Cencori.

Dashboard-first setup (recommended)

If you deployed an n8n Workflow agent in Cencori:

1. Open the agent's Configuration tab.
2. Add your n8n Base URL and n8n API Key.
3. Click Test Connection.
4. Click Install Starter Workflow to push a ready-made webhook workflow into n8n.
5. Use Publish / Unpublish to control whether the starter workflow is live.
6. Use Refresh Health to sync execution status and runtime metrics back into Cencori.

Fallback mode (no n8n API key)

If your n8n workspace plan does not expose API keys:

1. In Cencori n8n agent config, set Mode to Manual Fallback.
2. Save your n8n Base URL (API key is not required in this mode).
3. Import the starter JSON manually in n8n:
   • /examples/n8n/cencori-chat-webhook.json
4. Publish and manage the workflow directly in n8n UI.

In manual mode, one-click install/publish and execution health sync are disabled by design.

Quick start (manual node setup)

1. Add an HTTP Request node to your workflow.
2. In the node settings, configure the following:
   • Method: Select POST.
   • URL: Enter https://api.cencori.com/v1/chat/completions.
   • Authentication: Select Generic Credential Type.
   • Generic Credential Type: Select Header Auth.
3. Create a new Credential:
   • Name: Authorization
   • Value: Bearer cake_your_api_key_here (replace with your actual API key).
4. Under Body Parameters, set:
   • Send Body: Toggle to On.
   • Body Content Type: Select JSON.
   • Specify Body: Select Using JSON.
5. Paste the following JSON into the Body field:

   {
     "model": "gpt-4o",
     "messages": [
       {
         "role": "user",
         "content": "Hello world"
       }
     ]
   }

   (Tip: You can use expressions to insert dynamic data from previous nodes into the "content" field.)

cURL quick test (direct to Cencori)

Use this to validate your Cencori key outside n8n:

curl -sS https://api.cencori.com/v1/chat/completions \
  -H "Authorization: Bearer cake_your_api_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role":"user","content":"Hello from curl"}]
  }'

cURL create workflow in n8n (optional)

You can also create the starter workflow directly via n8n API:

curl -sS -X POST "https://your-n8n.example.com/api/v1/workflows" \
  -H "X-N8N-API-KEY: n8n_api_key_here" \
  -H "Content-Type: application/json" \
  --data-binary @examples/n8n/cencori-chat-webhook.json

Importable workflow template

You can import a ready-made n8n workflow template here:

• /examples/n8n/cencori-chat-webhook.json

The template includes:

• Webhook trigger node
• Cencori chat completion request node
• Structured response node for downstream steps

n8n hardening notes

• Keep baseUrl as https://api.cencori.com/v1 and call /chat/completions from n8n.
• Use raw model IDs like gemini-2.5-flash or llama-3.3-70b-versatile.
• CENCORI_AGENT_ID is optional for OpenAI-compatible clients. Cencori can derive the agent from the API key.

Use the native HTTP app to make requests to Cencori.

Step-by-Step Guide

1. Add a new module and search for HTTP.
2. Select the Make a request action.
3. Configure the module as follows:
   • URL: https://api.cencori.com/v1/chat/completions
   • Method: POST
   • Headers: Add a new header:
     • Name: Authorization
     • Value: Bearer cake_your_api_key_here
   • Body type: Raw
   • Content type: JSON (application/json)
4. In the Request content field, enter your JSON payload:

   {
     "model": "gpt-4o",
     "messages": [
       {
         "role": "user",
         "content": "{{1.text}}" 
       }
     ]
   }

   (Note: Replace {{1.text}} with the actual data pill from your trigger module.)
5. Check Parse response to automatically handle the JSON output.

Use the Webhooks by Zapier action to connect Cencori to thousands of apps.

Step-by-Step Guide

1. Add a step to your Zap and search for Webhooks by Zapier.
2. Select Custom Request as the event.
3. Configure the action:
   • Method: POST
   • URL: https://api.cencori.com/v1/chat/completions
   • Data Pass-Through?: No
   • Data:

     {
       "model": "gpt-4o",
       "messages": [
         {
           "role": "user",
           "content": "Summarize this: {{step_1_output}}"
         }
       ]
     }

   • Headers:
     • Content-Type: application/json
     • Authorization: Bearer cake_your_api_key_here
4. Test the step to see the AI response.

[!TIP] Why use Cencori for automation? Instead of managing individual OpenAI/Anthropic keys in every Zap, use a single Cencori key. This gives you centralized logs of all your automation AI usage.