rnnve Skills MCP
Modular Model Context Protocol host. Skills are TypeScript folders under skills/; every enabled skill’s tools appear on one Streamable HTTP endpoint.
This file is a Cursor/agent skill (when/how to use the deployed MCP). Do not confuse it with folders under skills/ — those are MCP skill modules registered in lib/registry.ts.
When to use
- Call Project SEKAI (
pjsk_*) tools for cards, characters, events, music, gacha, or counts - Connect a client (Cursor, Claude, etc.) to this MCP host
- Inventory what’s enabled (
skills_list) - Add or register a new MCP skill module in this repo
Endpoints
| Path | Role |
|---|---|
GET / |
Status page |
GET/POST /api/mcp |
MCP Streamable HTTP |
GET/POST /mcp |
Alias |
GET /api/health |
Health + skill list |
Protocol: JSON-RPC 2.0 (initialize → tools/list → tools/call). Stateless. CORS open.
Connect a client
Point a Streamable HTTP MCP client at the deployment (or local) URL:
{
"mcpServers": {
"skills": {
"url": "https://YOUR-DEPLOYMENT.vercel.app/api/mcp"
}
}
}
Replace YOUR-DEPLOYMENT with the Vercel project hostname for skills-mcp.
Tools (Project SEKAI)
| Tool | Use |
|---|---|
skills_list |
Host inventory |
pjsk_search |
Mixed query (e.g. kohane 4 cute) |
pjsk_card |
ID or filters |
pjsk_character |
Profile + rarity counts |
pjsk_event |
Current / id / name |
pjsk_music |
Songs |
pjsk_gacha |
Banners |
pjsk_counts |
Every character |
server defaults to jp. Also: en, tw, kr, cn.
Data is backed by the same master JSON sekai.best loads. First pjsk_card on a cold instance may be slow while cards.json is fetched and slimmed; later calls use memory + fetch cache.
Local
npm install
npm run dev
# POST http://localhost:3000/api/mcp
Add an MCP skill module
- Copy
skills/_templatetoskills/<name>. - Export a
Skill(id,title,description,version,tools[]). - Each tool:
name,description, JSON SchemainputSchema,handler(args). - Register in
lib/registry.ts. - Push to
main(Vercel rebuilds).
Keep handlers small and return JSON; the host wraps results as MCP text content.