---
title: "SUZU Monitor agent guide"
description: "Machine interfaces, authentication, request policy, and rate limits."
canonical: "https://global.suzu.party/agents.md"
---

# SUZU Monitor — Agent Guide

> How AI agents should work with SUZU Monitor: machine surfaces, authentication, crawl policy, rate limits, and discovery endpoints. Prefer the structured surfaces below over scraping the HTML dashboard — the dashboard is a WebGL SPA and yields nothing useful to a text parser.

SUZU Monitor is a real-time global intelligence dashboard: curated news feeds, a shared map-layer catalog, concrete panel implementations, country risk/resilience scores, AI briefs, forecasts, and market/supply-chain correlation, served as machine-readable JSON with documented methodology and provenance.

## Machine surfaces (use these)

- **MCP server (recommended):** `https://global.suzu.party/mcp` — Streamable HTTP; issue `tools/list` for the live inventory. Server card: https://global.suzu.party/.well-known/mcp/server-card.json
- **WebMCP (visible browser tabs):** https://global.suzu.party/docs/webmcp — experimental, page-local tools that operate the current homepage or dashboard UI in Chrome. WebMCP does not replace the hosted MCP server; use the server for remote, background, headless, or direct-data agents.
- **Docs MCP server:** `https://global.suzu.party/docs/mcp` — Streamable HTTP, public (no auth); search-and-retrieval tools over the documentation. Use it for "how do I…" questions; use the product MCP above for live data.
- **REST API:** base `https://api.worldmonitor.app` — OpenAPI spec: https://global.suzu.party/openapi.yaml (JSON: /openapi.json) · API catalog: https://global.suzu.party/.well-known/api-catalog
- **NLWeb:** `POST https://global.suzu.party/ask` (supports SSE) for natural-language questions; machine-readable dashboard view at `https://global.suzu.party/?mode=agent`
- **Agent Skills:** discovery index at https://global.suzu.party/.well-known/agent-skills/index.json · install via `npx skills add ziruiya/suzuglobal` (https://skills.sh/ziruiya/suzuglobal)
- **Agent setup page:** https://global.suzu.party/agent/ — per-client MCP setup for Claude, ChatGPT, Cursor, Gemini CLI, Grok Bot and generic MCP clients, plus skill install commands
- **Agent Plugin metadata:** https://global.suzu.party/plugin.json — public metadata for the Agent Plugins 1.0.0 repository package. Install the package from https://github.com/ziruiya/suzuglobal (root `plugin.json`, `mcp.json`, and `skills/*/SKILL.md` live in the repository, not as sibling HTTP files)
- **CLI:** `npx worldmonitor tools` lists every tool (public, no key) — https://www.npmjs.com/package/worldmonitor
- **SDKs:** Python `pip install worldmonitor-sdk` · Ruby `gem install worldmonitor` · Go `go get github.com/ziruiya/suzuglobal/sdk/go` · JavaScript npm `worldmonitor` — guide: https://global.suzu.party/docs/sdks
- **Sandbox / test environment:** https://global.suzu.party/sandbox/index.json — deterministic, schema-valid sample responses for representative REST operations; no auth, no quota, safe for CI. Guide: https://global.suzu.party/docs/sandbox
- **LLM briefings:** https://global.suzu.party/llms.txt (overview) · https://global.suzu.party/llms-full.txt (full reference) · section files: https://global.suzu.party/api/llms.txt (API) · https://global.suzu.party/docs/llms.txt (docs) · https://global.suzu.party/developers/llms.txt (developer portal) · https://global.suzu.party/blog/llms.txt (blog)
- **Schema map:** https://global.suzu.party/schemamap.xml — NLWeb schemamap indexing the structured-data surfaces
- **Research reports:** https://global.suzu.party/research/ — original source-backed research with downloadable CSV/JSON data, per-figure provenance, and stable citation URLs (no auth, no JavaScript required)
- **Developer portal:** https://global.suzu.party/developers.md — links every developer resource by name. Named resource pages: [MCP Server](https://global.suzu.party/mcp-server.md) · [OpenAPI Specification](https://global.suzu.party/openapi.md) · [SDKs](https://global.suzu.party/sdks.md)
- **Brand identity:** https://global.suzu.party/world-monitor.md — official name, canonical domain, NAP, press mentions
- **REST versioning and deprecation:** https://global.suzu.party/api-versioning.md — URL versioning, sunset timeline, Deprecation / Sunset / Link header contract (HTML: https://global.suzu.party/docs/api-versioning)

## Authentication

- **Anonymous** works for discovery endpoints, `tools/list`, and public data (world brief, product catalog, story pages). For MCP data tools, `get_sources` is the sole credential-free, daily-quota-free exception and has a separate fail-closed limit of 10 anonymous calls/minute/IP.
- **API key:** header `X-WorldMonitor-Key: wm_<40-hex>` for subscription-gated REST and MCP data calls — issue one at https://global.suzu.party/pro. All MCP data tools other than `get_sources` require subscription access. Full agent walkthrough: https://global.suzu.party/auth.md
- **OAuth2** for MCP (`scope=mcp`), with dynamic client registration at `/oauth/register`. Details in auth.md.

## Crawl & content-usage policy

- **robots.txt** (https://global.suzu.party/robots.txt): AI search, training and assistant agents (OAI-SearchBot, GPTBot, ChatGPT-User, Claude-SearchBot, ClaudeBot, Claude-User, PerplexityBot, Perplexity-User, Google-Extended, Applebot-Extended, DuckAssistBot, MistralAI-User) are explicitly allowed; bulk scrapers with no citation path (CCBot, Bytespider) are disallowed. Other crawlers follow the `User-agent: *` rules. `/api/` is off-limits to crawlers except the allowlisted story/OG/llms.txt/product-catalog routes.
- **Content-Signal:** `ai-train=yes, search=yes, ai-input=yes` — declared as a robots.txt group directive and as an origin-wide HTTP response header. Search indexing, assistant grounding/citation and model training are all welcome. Third-party data shown on SUZU Monitor keeps its upstream licence; this preference covers SUZU Monitor's own content.
- **User-Agent:** always send a descriptive `User-Agent` (e.g. `mytool/1.0 (+https://yoursite.example)`). Default HTTP-library UAs (`curl/*`, `python-requests/*`, empty strings) may get a 403 from the edge firewall — a 403 does NOT mean the endpoint is missing; retry with a real UA.

## Rate limits & plans

- Machine-readable pricing and plan limits: https://global.suzu.party/pricing.md · live JSON catalog: `GET https://global.suzu.party/api/product-catalog` (public, no key)
- Rate-limit documentation: https://global.suzu.party/docs/usage-rate-limits.md · auth matrix: https://global.suzu.party/docs/usage-auth
- Plan-limit responses include upgrade guidance; back off on 429 and honor `Retry-After`.

## Support & escalation

- https://global.suzu.party/support.md — support@suzu.party (general) · enterprise@suzu.party (sales)
- Status: https://global.suzu.party/status · Issues: https://github.com/ziruiya/suzuglobal/issues
- Source (AGPL-3.0): https://github.com/ziruiya/suzuglobal