extension-inference
caffeinelabsHosted LLM chat without API-key setup.
LLM integration
What it does
Guides Caffeine app development to use platform-hosted, authenticated LLM chat without collecting user API keys.
When to use it
Use whenever a Caffeine app needs an LLM, GPT, chatbot, or AI summarization.
How to use it
It changes how the agent implements LLM features in a Caffeine project.
What you provide
- An existing project
Uses
CI
Access · 1
This skill
Caffeine Inference
Write
Sends prompts to Caffeine Inference
Good to know
- may override your rules
What you need · 3
Requires Mops 2.13 or newer.
Requires the Caffeine-hosted chat endpoint that authenticates and bills requests for the app.
Config.fromEnv reads platform-provisioned inference credentials at runtime; missing credentials cause a trap.
About this skill
- Visibility
- Public
- Repository
- caffeinelabs/skills
- Created
- Oct 8, 2026
- Updated
- Oct 8, 2026
- Files
- 1