extension-openai
MANDATORY recipe for every Caffeine build that calls OpenAI (ChatGPT, GPT-4o, an LLM, a chatbot, embeddings). The ONLY supported path is the
Open source Repository Open in the app JSON README (API)
About
MANDATORY recipe for every Caffeine build that calls OpenAI (ChatGPT, GPT-4o, an LLM, a chatbot, embeddings). The ONLY supported path is the `openai-client` mops package with a canister-side API-key bearer. Hand-rolling `ic.http_request` to `api.openai.com/v1/...` is a FORBIDDEN anti-pattern — it leaks the bearer across replicated outcalls (security + 13× billing impact), bypasses the typed request/response bindings, and forces hand-rolled JSON on a language with poor JSON support. Load this skill whenever the user, spec, or any prior task mentions ChatGPT, GPT (any version), OpenAI, an LLM, a chatbot, or embeddings — and BEFORE writing any code that touches `api.openai.com`.
Details
- Kind
- Agent skills
- Topic
- AI, RAG & memory
- Publisher
- caffeinelabs
- Origin
- skillssh
- Category
- ferramentas
- Forks
- 1
- Last push
- 2026-09-07T06:58:52Z
- Repository state
- ativo
- Language
- Motoko
- License
- Apache-2.0
- Added
- 2026-08-30 15:21:15
- Updated
- 2026-09-09 21:04:25
- Origin id
caffeinelabs/skills/extension-openai
README
# Caffeine AI Skills
**Agent-readable skill files for building on the Caffeine AI.**
Structured, agent-readable skill files for every platform extension. Your AI reads the skill. It builds correctly. No hallucinations.
---
Every skill includes:
| Section | Purpose |
| ------------------- | -------------------------------------------------------------- |
| **Frontmatter** | YAML metadata: name, description, version, package compatibility |
| **Overview** | One paragraph. What the component does. |
| **Backend** | Motoko module API and usage examples (`mo:caffeineai-*`) |
| **Frontend** | React hooks and TypeScript API (`@caffeineai/*`) |
| **Implementation** | Tested, copy-paste-correct code blocks |
Skills may include backend only, frontend only, or both — depending on the component.
## Skills
All skills live in `skills/extension-*/SKILL.md`. Each skill is a self-contained markdown file with YAML frontmatter.
| Skill | Description |
| --- | --- |
| [authorization](skills/extension-authorization/SKILL.md) | Authorization system with role-based access control |
| [camera](skills/extension-camera/SKILL.md) | Web-camera support |
| [core-infrastructure](skills/extension-core-infrastructure/SKILL.md) | Core infrastructure providing backend connection configuration, storage client, and React app entry point |
| [email](skills/extension-email/SKILL.md) | Support for sending service/transactional emails |
| [email-calendar-events](skills/extension-email-calendar-events/SKILL.md) | Support for organising events/meetings and sending invitations by email |
| [email-marketing](skills/extension-email-marketing/SKILL.md) | Send personalised marketing emails to subscribers with an unsubscribe link |
| [email-raw](skills/extension-email-raw/SKILL.md) | Send an email with multiple to, cc and bcc addresses |
| [email-verification](skills/extension-email-verification/SKILL.md) | Support for sending a verification email with a link to prove email ownership |
| [http-outcalls](skills/extension-http-outcalls/SKILL.md) | HTTP outcalls performed by the backend canister |
| [invite-links](skills/extension-invite-links/SKILL.md) | Invite-link / RSVP based access for guest responses |
| [object-storage](skills/extension-object-storage/SKILL.md) | General file/object storage with browser-cached HTTP URL access |
| [qr-code](skills/extension-qr-code/SKILL.md) | QR code scanner using the camera |
| [stripe](skills/extension-stripe/SKILL.md) | Payment support based on Stripe, supporting credit cards and debit cards |
| [user-approval](skills/extension-user-approval/SKILL.md) | Approval-based user management |
## Usage
### Install via CLI
Works with any agent that supports skills (Claude Code, Cursor, Windsurf, Copilot, and more):
```
npx skills add caffeinelabs/skills
```
Browse available skills, pick your agent, and install. See [skills.sh](https://skills.sh) for details.
### Fetch a Single Skill
Fetch a single skill directly and place it wherever your agent reads instructions from:
```bash
curl -sL https://raw.githubusercontent.com/caffeinelabs/skills/main/skills/extension-authorization/SKILL.md
```
Replace `extension-authorization` with the skill directory name from the table above.
### Discover All Skills
List all available skills programmatically:
```bash
curl -s https://api.github.com/repos/caffeinelabs/skills/contents/skills | \
jq -r '.[].name'
```
## Skill File Structure for Extensions
Each `SKILL.md` follows a consistent structure:
```yaml
---
name: extension-name
description: Short description of what the component does.
version: 0.1.1
compatibility:
mops:
caffeineai-component-name: "~0.1.0"
npm:
"@caffeineai/component-name": "~0.1.0"
---
```
- **name** — identifier used for discovery and package resolution
- **description** — one-line summary for agents to match user intent
- **version** — skill file version
- **compatibility** — exact mops and/or npm package versions the skill documents
Below the frontmatter, skills are organized into `# Backend` and `# Frontend` sections with module APIs, usage examples, and integration notes.
## Packages
Each extension might leverage reusable package in `packages/`:
```
packages/
<name>/
backend/ # Motoko package (mops.toml)
frontend/ # TypeScript package (package.json)
```
- **Backend** packages are published to the [mops](https://mops.one) registry as `caffeineai-*`
- **Frontend** packages are published to GitHub Packages as `@caffeineai/*`
## License
Apache-2.0