MCP Platform
Connect via MCP to Claude Desktop, Cursor, or any MCP client. Run strategic analyses programmatically with full gate coverage, research, and audit trail.
13 MCP tools · 97 gates · 11 research sources · API key in 60 seconds
| Tool | Credits | Async | Description |
|---|---|---|---|
| decision_start | 1 or 5 | Yes | Full strategic analysis. Auto-tier if tier omitted. |
| decision_status | 0 | No | Poll run status: phase, progress, gates fired |
| decision_result | 0 | No | Retrieve completed deliverable |
| quota_usage | 0 | No | Check remaining runs |
| convert_document | 0 | No | Base64 → Markdown (up to 50 MB) |
| convert_documents | 0 | No | Batch conversion (max 20) |
| ingest_from_url | 0 | No | Download + convert + ingest from URL |
| memory_ingest | 0 | No | Text → knowledge base (500 KB max) |
| memory_ingest_batch | 0 | No | Batch ingest (max 20) |
| memory_add | 0 | No | Add atomic fact to KB |
| memory_search | 0 | No | FTS5 + vector hybrid search |
| simulate_audience | 5 | Yes | Multi-round audience debate simulation |
| decision_audit_deep | 15 | Yes | 10-stage adversarial audit pipeline |
Pick a plan. Get your API key in 60 seconds.
Sign up at big4.cloud — you get an API key automatically on email verification.
Add the big4.cloud server to your MCP client config (see below).
Start asking strategic questions — the engine handles tier selection, research, and delivery.
Works with Claude Desktop, Cursor, Kiro, and any MCP-compatible client.