Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

Agent World

REMOTE · AGENTS.HOTSCHMOE.COM · SCANNED SEP 20

Search, URL-to-markdown, change detection, live data, x402 trust tools. USDC pay-per-call.

23 Trust /100
Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. How we score → Why this is hard to score →

Endpoint Security57
Transport & Reachability0
Schema Quality & AI Usability0
  • Schema not yet verified: we couldn't read the endpoint's schema.Unverified
Stability & Change Management0
  • Stability not yet verified: not enough scan history yet (needs a 30-day window).Unverified
Tool Coverage0
  • Tool coverage not yet verified: we couldn't read the endpoint's tools.Unverified
Tool Safety0
  • Tool safety not yet verified: we couldn't read the endpoint's tools.Unverified
Capabilities0
  • Capabilities not yet verified: we couldn't read the endpoint's capabilities.Unverified

Unverified: 5 categories

Categories scored 0 because we could not verify them: authentication we do not have, an unreachable endpoint, or not enough scan history. We only credit what we can confirm.

Install

How do I install the Agent World MCP server?

Agent World is a hosted endpoint at https://agents.hotschmoe.com/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

remote · agents.hotschmoe.com

# add to Claude Code
claude mcp add --transport http com-hotschmoe-agents-agentworld 'https://agents.hotschmoe.com/mcp'
// .cursor/mcp.json
{
  "mcpServers": {
    "com-hotschmoe-agents-agentworld": {
      "url": "https://agents.hotschmoe.com/mcp"
    }
  }
}
// .vscode/mcp.json
{
  "servers": {
    "com-hotschmoe-agents-agentworld": {
      "type": "http",
      "url": "https://agents.hotschmoe.com/mcp"
    }
  }
}
# ~/.codex/config.toml
[mcp_servers.com-hotschmoe-agents-agentworld]
url = "https://agents.hotschmoe.com/mcp"
// opencode.json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "com-hotschmoe-agents-agentworld": {
      "type": "remote",
      "url": "https://agents.hotschmoe.com/mcp",
      "enabled": true
    }
  }
}
# add to OpenClaw
openclaw mcp add com-hotschmoe-agents-agentworld --url 'https://agents.hotschmoe.com/mcp' --transport streamable-http
# ~/.hermes/config.yaml
mcp_servers:
  com-hotschmoe-agents-agentworld:
    url: "https://agents.hotschmoe.com/mcp"
// ~/.netclaw/config/netclaw.json
{
  "McpServers": {
    "com-hotschmoe-agents-agentworld": {
      "Transport": "http",
      "Url": "https://agents.hotschmoe.com/mcp"
    }
  }
}
# add to Vellum
assistant mcp add com-hotschmoe-agents-agentworld -t streamable-http -u 'https://agents.hotschmoe.com/mcp'
// mcp.json
{
  "mcpServers": {
    "com-hotschmoe-agents-agentworld": {
      "type": "http",
      "url": "https://agents.hotschmoe.com/mcp"
    }
  }
}

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 16 Sept 26 −41
    • Endpoint reachability: reachable → not serving MCP security
    • Tool safety: pass → unverified security
    • Stability: 0.03 → unverified security
    • Transport: pass → fail security
    • Authorization: Authorisation not fully verified: no authorisation is required to connect, but we couldn't read the tool list to see what that exposes. security
    • Capabilities: pass → unverified functional
    • Tool coverage: 100 → unverified functional
    • First check of Schema quality: unverified functional
  • 15 Sept 26 0
    • Schema quality: 2492 → 2848 functional
    • Stability: unverified → 0.03 functional
    • Server version: 1.0.9 → 1.0.10 functional
    • New tool “world_diff” functional
    • New tool “world_diff_lite” functional
  • 14 Sept 26 64

    First indexed and scored.

Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 20 Sept 2026 · Probed https://agents.hotschmoe.com/mcp

TLS valid

Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=agents.hotschmoe.com CN=YR2,O=Let's Encrypt,C=US 7 Sept 2026 6 Dec 2026 RSA 4096 SHA256-RSA 6c209871dc5907025d4a9ee8c9333e69f31
SANs: agents.hotschmoe.com, agentsworld.hotschmoe.com
CN=YR2,O=Let's Encrypt,C=US (CA) CN=Root YR,O=ISRG,C=US 3 Sept 2025 2 Sept 2028 RSA 2048 SHA256-RSA 4ebd24947e24d394802d84a52fd5b319
CN=Root YR,O=ISRG,C=US (CA) CN=ISRG Root X1,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 RSA 4096 SHA256-RSA f24b6d17f9d9ad7cb1c9fea78782699f

Background: What to check on a remote MCP endpoint →

DNSSEC insecure

Validation of agents.hotschmoe.com. Not signed

Zone DS Keys Algorithms Outcome
. trust_anchor 20326, 38696 8, 8 Verified
com. present 19718 13 Verified
hotschmoe.com. absent Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation
Authentication No authorisation required

The endpoint answered without asking for a token. Anyone who knows the URL can reach it.

Result No authorisation required
HTTP status 405
Header Value
x-content-type-options nosniff

Background: How OAuth 2.1 works in the 2026 MCP spec →

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://agents.hotschmoe.com/mcp HTTP error 405
http (plaintext) http://agents.hotschmoe.com/mcp HTTPS enforced 301 https://agents.hotschmoe.com/mcp
MCP tools · 17 exposed · ~2,593 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
bazaar_pulse ~161

The x402 agent economy in one call: PayAI Bazaar market vitals (28K+ resources, 1886 hosts, settlement volume 24h/7d/30d, network mix, 24h uptime, avg latency), the venue's own top-20 merchants by settlements with volume, the world's accumulating trend history (30-min snapshots, deltas), and the world's position in the catalog. The only tool on this server about the MARKET rather than about your endpoint. Set lite:true for market vitals + top 5 merchants. [paid: 0.0050 USDC on Base mainnet via x402]

NameTypeReqDescription
litebooleanIf true: market vitals + top 5 merchants only (lighter response)

No output schema declared.

No examples provided.

live_search ~146

Search the world's LIVE INDEX: the only x402 directory that lists resources that actually WORK — every row has a fresh dated signed liveness receipt AND a Base-mainnet payment option (out of a 28K-listing catalog where ~84% measure dead). Substring search over name, description, origin. Returns up to 50 rows with price, networks, last-verified timestamp, and the verifiable receipt URL for each. [paid: 0.0010 USDC on Base mainnet via x402]

NameTypeReqDescription
limitintegerMax rows, 1-50 (default 20)
querystringKeyword to search (empty = top rows)

No output schema declared.

No examples provided.

world_catalog ~123

FREE catalog of everything this world sells: every paid x402 data product (digest, sentinel, pingspot, world-state, attest, audit, knock, bazaar, live-index, fleet, pulse) plus OpenAI-compatible LLM inference, each with its public URL, USDC price on Base mainnet, and a one-line description. Built live from the gateway's OpenAPI, so it never drifts. Use it to discover what is available before paying for a specific tool; the same catalog is the world's $0 PayAI Bazaar listing. [free]

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

world_diff ~237

Change detection for any public URL: the world fetches the target, hashes the raw body (SHA-256), and compares against a baseline — the world's own stored baseline for that URL (kept across your calls; a failed probe never destroys it) or your own baseline sha256 from an earlier world_knock/world_attest/world_diff. Returns a signed (Ed25519, kid world-diff) verdict — unchanged, changed, first-observation, or target-error — with a unified line diff of exactly what changed (text targets; binary bodies report byte counts). The pull-based monitoring product: no webhook, no 24h expiry, one recurring parameter. 0.001 USDC on Base mainnet via x402. [paid: 0.0010 USDC on Base mainnet via x402]

NameTypeReqDescription
baselinestringOptional 64-char sha256 hex baseline (omit to use the world's stored baseline for this URL)
maxintegerMax diff chars, 100-20000 (default 4000)
urlstringyesPublic http(s) URL to check for changes

No output schema declared.

No examples provided.

world_diff_lite ~119

FREE preview of world_diff: the change-detection verdict (unchanged/changed/first-observation/target-error) plus both SHA-256 hashes for any public URL (5/hour per IP; read-only on the world's baseline store). Use it to check whether a page changed before the paid diff; the paid world_diff (0.001 USDC) returns the unified line diff of exactly what changed, with the Ed25519-signed dated record. [free]

NameTypeReqDescription
urlstringyesPublic http(s) URL to check

No output schema declared.

No examples provided.

world_digest ~82

Live 30-minute snapshot: BTC/ETH/SOL spot prices, USD/EUR/JPY FX, weather, Hacker News top 10, GitHub trending repos. One call, no parameters, JSON. For market context, price checks, news digests, dashboard data. [paid: 0.0001 USDC on Base mainnet via x402]

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

world_doctor ~131

FREE x402 protocol-conformance report for any public endpoint: the world's doctor scores it 0-100 across 13 checks (402 challenge decodability, accept validation, discovery docs, advertised-vs-challenged price, CORS, resource.url) and EIP-191-signs the report with the world wallet (verify in two lines). The buyer-side safety check before wiring a client to an x402 merchant. Same engine as the public /x402/doctor route. [free]

NameTypeReqDescription
targetstringyesPublic http(s) origin or URL of an x402 endpoint to check

No output schema declared.

No examples provided.

world_doctor_fleet ~170

Batch x402 conformance audit: up to 10 origins in ONE call — each gets a signed 0-100 conformance report (13 checks: challenge decodability, accept validation, discovery surfaces, advertised-vs-challenged price, CORS) plus a fleet summary (by-verdict counts, average, worst offender). 0.005 USDC on Base mainnet via x402. One call instead of ten; not subject to the free /doctor 5/hour limit. For vetting a batch of x402 merchants before wiring a client to any of them. [paid: 0.0050 USDC on Base mainnet via x402]

NameTypeReqDescription
originsstringyesComma-separated public http(s) origins or URLs to check (1-10)

No output schema declared.

No examples provided.

world_fleet ~174

Batch liveness audit: up to 25 public URLs in ONE call. Per URL: dated 3-state verdict (up/degraded/down), final HTTP status, latency, SHA-256 body hash, x402-challenge detection (decodable? which networks? mainnet accepts?). Plus a fleet summary (up/degraded/down, dead ratio, x402-challenged, mainnet-payable). 25 world_knock calls cost 0.025; this costs 0.005. For directory re-knock sweeps, dead-listing cleanup, and vetting a batch of endpoints before paying any of them. [paid: 0.0050 USDC on Base mainnet via x402]

NameTypeReqDescription
urlsstringyesComma-separated public http(s) URLs to knock (1-25)

No output schema declared.

No examples provided.

world_knock ~133

Dated liveness receipt for any public URL: Ed25519-signed, timestamped record of final HTTP status, latency, SHA-256 body hash, x402-challenge detection (decodable? which networks? mainnet accepts?), and a 3-state verdict (up/degraded/down). Published at a durable public URL and verifiable against the world-knock public key. Use to prove a URL answered (or not) at a given time. [paid: 0.0010 USDC on Base mainnet via x402]

NameTypeReqDescription
urlstringyesPublic http(s) URL to knock

No output schema declared.

No examples provided.

world_read ~309

Fetch any URL as clean markdown for LLM/agent context (web page to markdown for your LLM): reads any public URL and returns markdown (headings, links, lists, tables, code; scripts/styles stripped) trimmed to max chars (default 8000, cap 50000), plus title, meta description, canonical, JSON-LD types, link stats, SHA-256 of the raw body, and an Ed25519-signed dated record (kid world-read) of exactly what the world saw. JSON and plain-text targets return as-is with stats. The action tool: it does work for you instead of reporting on endpoints. 0.001 USDC on Base mainnet via x402. Also listed in the agent402 Smart Order Router's open index: to buy through the router instead, POST https://agent402.tools/api/route/execute with {"task": "fetch <url> as markdown", "include": "external"} plus a $0.01 USDC x402 payment on Solana — the router resolves the task against the open index and pays the winning seller from its own wallet (we receive $0.001; the extra $0.009 is the router's fee). [paid: 0.0010 USDC on Base mainnet via x402]

NameTypeReqDescription
maxintegerMax chars to return, 200-50000 (default 8000)
urlstringyesPublic http(s) URL to read

No output schema declared.

No examples provided.

world_read_lite ~109

FREE preview of world_read: the first 500 chars of the clean-markdown render of any public URL, with title and content type (5/hour per IP). Use it to check whether a page is worth the paid read; the paid world_read (0.001 USDC) returns up to 50000 chars with JSON-LD, link stats, and the Ed25519-signed dated record. [free]

NameTypeReqDescription
urlstringyesPublic http(s) URL to read

No output schema declared.

No examples provided.

world_search ~322

Web search for agents: any query returns the top 10 web results (rank, title, URL, snippet) from DuckDuckGo (Bing fallback) — no API key, no account — optionally with the top result already fetched as clean markdown for LLM context (fetch:true, max chars default 8000), plus an Ed25519-signed dated record (kid world-search) of exactly what the world saw. The search half of the search-then-read loop: find sources with world_search, digest them with world_read. 0.002 USDC on Base mainnet via x402. Also listed in the agent402 Smart Order Router's open index: to buy through the router instead, POST https://agent402.tools/api/route/execute with {"task": "<your query>", "include": "external"} plus a $0.01 USDC x402 payment on Solana — the router resolves the task against the open index and pays the winning seller from its own wallet (we receive $0.002; the extra $0.009 is the router's fee). [paid: 0.0020 USDC on Base mainnet via x402]

NameTypeReqDescription
fetchbooleanIf true: also return the top result as clean markdown
maxintegerMax chars for the fetched top result, 200-50000 (default 8000)
nintegerMax results, 1-25 (default 10)
qstringyesSearch query (1-300 chars)

No output schema declared.

No examples provided.

world_search_lite ~103

FREE preview of world_search: the top 3 web results (title, URL, snippet) for any query (5/hour per IP). Use it to check what a query returns before the paid search; the paid world_search (0.002 USDC) returns up to 25 results, optional top-result-as-markdown, and the Ed25519-signed dated record. [free]

NameTypeReqDescription
qstringyesSearch query (1-300 chars)

No output schema declared.

No examples provided.

world_sentinel ~65

Uptime status of 12 monitored HTTPS hosts: 3-state health (up/degraded/down) per host, recent incidents, last-change times. For dependency checks and status-page data. [paid: 0.0001 USDC on Base mainnet via x402]

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

world_status ~64

FREE health check of the agent world that serves this MCP endpoint: digest feed freshness, sentinel 3-state host health summary, live citizen count, and the world's payment wallet. Free at any time — use it to confirm the world is alive before paying for data. [free]

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

world_watch ~145

FREE status of the world's watch product: latest state of any ACTIVE continuous-monitoring watch on the given URL (verdict up/degraded/down, last check, checks count, expiry) — the world knocks registered targets every ~30 min and pushes Ed25519-signed alerts to buyer webhooks on state change. Registration is paid: GET /x402/watch?url=<url>[&callback=<webhook>] (0.002 USDC on Base mainnet, up to 24h). If no watch is active on the URL you get a registration pointer. [free]

NameTypeReqDescription
urlstringyesAbsolute http(s) URL to look up in the active watch set

No output schema declared.

No examples provided.

Common questions

What is the Agent World MCP server?

Agent World is an MCP server listed in the public MCP registry as com.hotschmoe.agents/agentworld. Search, URL-to-markdown, change detection, live data, x402 trust tools. USDC pay-per-call. This page covers its hosted endpoint (https://agents.hotschmoe.com/mcp).

Is the Agent World MCP server safe to use?

Agent World scores 23 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the Agent World MCP server expose?

Agent World exposes 17 tools: world_digest, world_sentinel, live_search, world_knock, world_read, and 12 more. Their descriptions and schemas cost roughly 2,593 tokens of context every time the server is loaded.

Does the Agent World MCP server require authentication?

No. We connected to Agent World without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

Is the Agent World MCP server still maintained?

Agent World is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.