API Documentation
Everything you need to integrate supaskills into your Claude Code setup.
Quick start
1. Sign up and generate an API key from your dashboard settings.
2. Use your key with the REST API or configure MCP access.
Authentication
All API requests require a Bearer token in the Authorization header.
API keys are prefixed with sk_supa_ and are SHA-256 hashed before storage. Keep your keys secure.
Authorization: Bearer sk_supa_a1b2c3d4e5f6...Endpoints
/api/v1/skillsList and search published skills.
Query parameters
qFull-text search querycategoryFilter by category slugdomainFilter by domain nametierFilter by tier (diamond, platinum, gold, silver)sortSort by: score, rating, newestpagePage number (default: 1)limitItems per page (default: 20, max: 100)languageFilter by language (de, en)/api/v1/skills/:slugActivate a skill by slug. Returns the full system prompt and metadata.
Query parameters
versionPin to a specific version (e.g., 1.0.0). Default: latest./api/v1/categoriesList all skill categories with skill counts.
/api/v1/skills/:slug/feedbackSubmit a rating (1\u20135) and optional text feedback.
{ "rating": 5, "text": "Excellent skill!" }/api/v1/powerpacks/:slugActivate a PowerPack bundle. Pro and Max plans only.
/api/v1/skills/:slug/activateActivate a skill. Consumes one slot for 7 days.
{ "activation_id": "...", "expires_at": "...", "slots_used": 2, "slots_max": 15 }/api/v1/skills/:slug/activateDeactivate a skill. Frees the slot immediately.
{ "deactivated": true, "slots_used": 1, "slots_max": 15 }/api/v1/me/activationsList your active skill slots.
{ "slots_used": 2, "slots_max": 15, "plan": "pro", "activations": [...] }MCP Integration
supaskills.ai exposes a native MCP (Model Context Protocol) endpoint for Claude Code. Add it to your Claude Code settings:
Available MCP tools
search_skillsSearch and filter the skill catalogload_skillActivate a skill and get its system prompt by slugload_powerpackActivate a PowerPack bundlelist_categoriesList all skill categoriessuggest_skillGet a skill recommendation for a taskResponse format
All responses follow a consistent JSON structure:
// Success
{ "data": { ... }, "meta": { "page": 1, "limit": 20, "total": 42 } }
// Error
{ "error": { "code": "RATE_LIMIT_EXCEEDED", "message": "...", "status": 429 } }Rate limits
| Plan | Active Slots | Per-minute |
|---|---|---|
| Free | 3 | 2 |
| Pro | 15 | 15 |
| Max | 100 | 30 |