Connect AI Coach to your agent
Add AI Coach as an MCP server in 30 seconds. Your agent gets search_skills, get_skill, install_skill, search_marketplace, list_models, and compare_models — all free, no key required for v1.
Endpoint
https://aicoach.pw/api/mcp/v1 Transport
streamable-http Auth (v1)
none for reads · bearer optional Claude Code
Add to your ~/.claude/mcp.json or per-project .claude/mcp.json:
{
"mcpServers": {
"aicoach": {
"url": "https://aicoach.pw/api/mcp/v1",
"transport": "http"
}
}
}Or with an API key (later, when paid tools land):
{
"mcpServers": {
"aicoach": {
"url": "https://aicoach.pw/api/mcp/v1",
"transport": "http",
"headers": {
"Authorization": "Bearer mcp_YOUR_KEY_HERE"
}
}
}
}Cursor
Add to your .cursor/mcp.json:
{
"mcpServers": {
"aicoach": {
"url": "https://aicoach.pw/api/mcp/v1"
}
}
}Codex CLI
Add to your ~/.codex/config.toml:
[mcp_servers.aicoach]
type = "http"
url = "https://aicoach.pw/api/mcp/v1"
headers = { Authorization = "Bearer mcp_YOUR_KEY_HERE" }Zed
Add to your Zed settings.json:
{
"context_servers": {
"aicoach": {
"transport": "http",
"url": "https://aicoach.pw/api/mcp/v1"
}
}
}Test it
From any terminal:
curl -sX POST https://aicoach.pw/api/mcp/v1 \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'API key (optional today, required for paid tools later)
v1 is fully read-only and works without auth. When paid tools land (running nano apps, model bake-offs), you'll attach a Bearer token to authenticate the calling agent against your AI Coach wallet.
Sign in to issue a key →