CrewEvo Developer Platform
Call CrewEvo from your signed-in applications and scripts using a chat-completion JSON format. Authentication is required; response times depend on server availability and load.
Uses the configured AI nodes and bounded fallback attempts. Model selection is managed by the server.
Accepts messages and returns choices. Streaming, arbitrary model selection and full OpenAI API compatibility are not implemented.
Request web or GitHub context with connectors plus webUrl or repoUrl. Access and results depend on connector availability.
Authentication
POST requires Authorization: Bearer followed by a Firebase ID token issued for your signed-in CrewEvo account. A Firebase project API key is not an ID token. Previous anonymous requests and placeholder keys now receive HTTP 401.
Use your Firebase client SDK to obtain and refresh the token. On 401, refresh your session or sign in again. Never commit, log or share tokens. The examples assume an authenticated session; they do not create one.
Quick Start Code Examples
# Set CREWEVO_ID_TOKEN securely to a current Firebase ID token.
curl -X POST https://crewevo.com/api/v1/chat \
-H "Authorization: Bearer $CREWEVO_ID_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"messages": [
{"role": "user", "content": "Suggest three launch actions for my coffee shop."}
],
"locale": "en"
}'Endpoint Reference
| Method | Endpoint | Description |
|---|---|---|
| POST | /api/v1/chat | Generate chat responses; requires a Firebase ID token. |
| GET | /api/v1/chat | Read API metadata; this does not probe model health. |