MCP server
Every primitive is auto-published as an MCP tool. One endpoint, every action.
What it is
AIARCO runs a Model Context Protocol server at https://api.asc.aiarco.com/v1/mcp. Every primitive becomes a tool the agent can call (pods.run, storage.upload, edge.waf.rules.add, etc.) with the same scope checks as the REST API. Plug it into Claude Desktop, Cursor, OpenAI Agents or any MCP-compatible client.
When to use it
- Give an AI agent the ability to deploy your own infra.
- Use in-IDE assistants for ad-hoc ops.
- Build your own MCP-native product on top of AIARCO.
Quickstart
# Discover the tool registry
curl https://api.asc.aiarco.com/v1/mcp/registry -H 'Authorization: Bearer asc_live_…'# Configure in mcp.json:
# { "servers": { "aiarco": { "url": "https://api.asc.aiarco.com/v1/mcp", "headers": { "Authorization": "Bearer asc_live_…" } } } }// Same as Python — config-driven through the host's mcp.json or equivalent.Limits & quotas
| Limit | Default | Burst | Notes |
|---|---|---|---|
| Tools exposed | All primitives | — | Auto-generated |
| Auth | Bearer asc_… | — | |
| Per-tool scope check | Yes | — |
Pricing
See pricing. Pay-as-you-go, billed monthly via Stripe.
API surface
POST /v1/mcp(RPC)GET /v1/mcp/registry
Required scope(s): mcp:invoke. See Scopes.
Security
All access is authenticated and scoped. See Auth & scopes and Network controls.