Skip to content

ThirdMade Shelf MCP server

net.thirdmade/shelfthirdmade.netstreamable-httpv3.1.0Scanned

26 pay-per-call tools for agents: scrape, verify, guards, crypto. x402 USDC on Base.

26 tools: 18 change things and 8 only read.

Tools: 0 delete, 0 irreversible, 18 write, 8 read.
  • Reads private data · 0
  • Reads text others write · 4
  • Can send data out · 6
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

18 tools here change things. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the ThirdMade Shelf MCP server can do

The ThirdMade Shelf MCP server is published by thirdmade.net, a domain the MCP Registry verified. It has 26 tools. It can create or change data (18 tools: robots, page_meta, domain_facts, link_graph and 14 more), but cannot delete data. 8 tools only read.

4 tools bring in text other people can write (robots, llms_txt, scrape and 1 more); 6 tools can send data out of your system (robots, llms_txt, page_meta and 3 more).

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 18, 2026, version 3.1.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

26 tools

Changes things 18Claude Code: ask
  • captcha_evalClassifies captcha type (reCAPTCHA, hCaptcha, Turnstile) and difficulty.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Classifies captcha type (reCAPTCHA, hCaptcha, Turnstile) and difficulty. Eval only — we do NOT solve captchas. [price: $0.01/call USDC via x402]
  • chain_statsBlock height + chain stats for Base or Ethereum.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Block height + chain stats for Base or Ethereum. [price: $0.001/call USDC via x402]
  • domain_factsLive-checks a domain: is it up, does it have TLS, what's the title?No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Live-checks a domain: is it up, does it have TLS, what's the title? Quick domain health check for agents evaluating links. [price: $0.001/call USDC via x402]
  • gas_nowCurrent gas prices in Gwei for Base, Ethereum, or Arbitrum via public RPC.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Current gas prices in Gwei for Base, Ethereum, or Arbitrum via public RPC. [free]
  • geo_timeGeocode + timezone + currency context for any location string.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Geocode + timezone + currency context for any location string. [price: $0.001/call USDC via x402]
  • hallucination_checkEntailment scoring: does the provided context support this claim?No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Entailment scoring: does the provided context support this claim? Research agents use this to self-check outputs. [price: $0.001/call USDC via x402]
  • image_genGenerates 1 PNG from a text prompt (FLUX.1-dev, 512px).No findings

    Why it is in this group

    • The description's first verb is "generate": a word for a change.
    Description, as the server wrote it
    Generates 1 PNG from a text prompt (FLUX.1-dev, 512px). Async: returns job ID + poll URL. GPU-backed via RunPod ComfyUI, scale-to-zero. [price: $0.02/call USDC via x402]
  • jsonifyCoerces messy text (key=value lines, CSV-ish, sentences) into clean JSON.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Coerces messy text (key=value lines, CSV-ish, sentences) into clean JSON. [price: $0.001/call USDC via x402]
  • nonceCurrent transaction nonce for any address on Base or Ethereum.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Current transaction nonce for any address on Base or Ethereum. Agent tx bots need this before sending transactions. [price: $0.001/call USDC via x402]
  • page_metaExtracts OG tags, canonical URL, title, and meta description from any web page.Can send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Extracts OG tags, canonical URL, title, and meta description from any web page. Useful for link previews, SEO audits, and content enrichment. [price: $0.001/call USDC via x402]
  • pii_guardDetects PII (emails, phones, SSNs, credit cards) in any text with character spans.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Detects PII (emails, phones, SSNs, credit cards) in any text with character spans. Compliance agents use this before processing user data. [price: $0.001/call USDC via x402]
  • prompt_guardScores prompt-injection / jailbreak risk for LLM inputs.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Scores prompt-injection / jailbreak risk for LLM inputs. Agent builders use this to sanitize user inputs before hitting their LLM. [price: $0.01/call USDC via x402]
  • robotsParses robots.txt for any domain — shows crawl rules, disallowed paths, sitemaps.Reads text others writeCan send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe description's first sentence says "crawl": it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Parses robots.txt for any domain — shows crawl rules, disallowed paths, sitemaps. Agents use this to check if they're allowed to scrape a site before doing it. [free]
  • scrapeScrapes any public web page and returns clean structured JSON (title, text, length).Reads text others writeCan send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe name "scrape" says "scrape": it reads web pages, which anyone can write.
    • untrusted-inputThe description's first sentence says "scrape": it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Scrapes any public web page and returns clean structured JSON (title, text, length). The workhorse — agents use this for real data extraction. [price: $0.01/call USDC via x402]
  • sentimentSentiment classification (positive/negative/neutral) + emotion keyword hits.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Sentiment classification (positive/negative/neutral) + emotion keyword hits. [price: $0.001/call USDC via x402]
  • sweepOne verify-loop lane: confidence-labeled findings on any topic.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    One verify-loop lane: confidence-labeled findings on any topic. The flagship teaser — full sweeps at $0.50-5 ship when this crosses BUILD. [price: $0.05/call USDC via x402]
  • token_securityToken security heuristics: honeypot signals, proxy detection, LP data.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Token security heuristics: honeypot signals, proxy detection, LP data. Trading agents use this to avoid rugs. [price: $0.01/call USDC via x402]
Only reads 8Show 8 read-only tools
  • ens_lookupENS name resolution + reverse lookup via public Ethereum RPC.No findings

    Why it is in this group

    • The name "ens_lookup" says "lookup": a word for reading.
    Description, as the server wrote it
    ENS name resolution + reverse lookup via public Ethereum RPC. [price: $0.001/call USDC via x402]
  • entity_findExtracts named entities (people, companies, places) from text.No findings

    Why it is in this group

    • The name "entity_find" says "find": a word for reading.
    Description, as the server wrote it
    Extracts named entities (people, companies, places) from text. Wikidata resolution next iteration. [price: $0.001/call USDC via x402]
  • image_statusPoll image-gen job status by job ID.No findings

    Why it is in this group

    • The name "image_status" says "status": a word for reading.
    Description, as the server wrote it
    Poll image-gen job status by job ID. Returns image base64 PNG when COMPLETED. [free]
  • json_guardValidates JSON, detects PII inside values, reports schema stats.No findings

    Why it is in this group

    • The description's first verb is "validate": a word for reading.
    Description, as the server wrote it
    Validates JSON, detects PII inside values, reports schema stats. Pipeline agents use this as a data-quality gate. [price: $0.001/call USDC via x402]
  • llms_txtFetches and normalizes llms.txt / llms-full.txt from any site — the emerging standard for making websites readable to AI agents.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The description's first verb is "fetch": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "fetch" and "website": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Fetches and normalizes llms.txt / llms-full.txt from any site — the emerging standard for making websites readable to AI agents. [free]
  • search_verifyChecks if a URL supports a claim.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "search_verify" says "search": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "url": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Checks if a URL supports a claim. Returns confidence score. Research agents use this to verify sources. [price: $0.01/call USDC via x402]
  • similar_domainsFinds similar/competitor domains by name-similarity.No findings

    Why it is in this group

    • The description's first verb is "find": a word for reading.
    Description, as the server wrote it
    Finds similar/competitor domains by name-similarity. Lead-gen agents use this for prospecting. [price: $0.001/call USDC via x402]
  • summarizeExtractive summary (3 sentences) + top-5 keywords from any text.No findings

    Why it is in this group

    • The name "summarize" says "summarize": a word for reading.
    Description, as the server wrote it
    Extractive summary (3 sentences) + top-5 keywords from any text. [price: $0.001/call USDC via x402]

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write4
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out6
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 19:16 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
18
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 18 tools that change things (18 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://shelf.thirdmade.net/mcp: tools/list returned 26 tools to an anonymous client, including robots (write), page_meta (write), domain_facts (write), link_graph (write), scrape (write) and 13 more

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 3.1.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http shelf https://shelf.thirdmade.net/mcp
claude mcp add --transport http shelf https://shelf.thirdmade.net/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__shelf__entity_find",
      "mcp__shelf__similar_domains",
      "mcp__shelf__ens_lookup",
      "mcp__shelf__summarize",
      "mcp__shelf__image_status",
      "mcp__shelf__json_guard"
    ],
    "ask": [
      "mcp__shelf__robots",
      "mcp__shelf__llms_txt",
      "mcp__shelf__page_meta",
      "mcp__shelf__domain_facts",
      "mcp__shelf__link_graph",
      "mcp__shelf__scrape",
      "mcp__shelf__captcha_eval",
      "mcp__shelf__geo_time",
      "mcp__shelf__search_verify",
      "mcp__shelf__gas_now",
      "mcp__shelf__chain_stats",
      "mcp__shelf__nonce",
      "mcp__shelf__token_security",
      "mcp__shelf__jsonify",
      "mcp__shelf__sentiment",
      "mcp__shelf__image_gen",
      "mcp__shelf__pii_guard",
      "mcp__shelf__prompt_guard",
      "mcp__shelf__hallucination_check",
      "mcp__shelf__sweep"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as shelf. Review it before you use it.

Capability tags behind 6 rules
  • mcp__shelf__robots untrusted-input, outbound
  • mcp__shelf__llms_txt untrusted-input, outbound
  • mcp__shelf__page_meta outbound
  • mcp__shelf__link_graph outbound
  • mcp__shelf__scrape untrusted-input, outbound
  • mcp__shelf__search_verify untrusted-input, outbound

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/net.thirdmade/shelf/verify?hash=b490d120e1115552d33f4596d19df468d64ff752364fad9b32c38054b03ca404" | jq -e '.changed == false'

Pinned hash b490d120e111, from Nov 18, 2026, 21:17 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/net.thirdmade/shelf/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for net.thirdmade/shelf
[![Anlyon Radar](https://radar.anlyon.com/badge/net.thirdmade/shelf.svg)](https://radar.anlyon.com/servers/net.thirdmade/shelf)

Source: MCP Registry. Synced Oct 21, 2026, 18:53 UTC.

ThirdMade Shelf MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the ThirdMade Shelf MCP server safe?

No scan can promise that. What its definitions show: 26 tools, 18 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the ThirdMade Shelf MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the ThirdMade Shelf MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the ThirdMade Shelf MCP server have?

26 tools: 18 write, 8 read. robots, page_meta, domain_facts, link_graph, scrape, captcha_eval, geo_time, gas_now and 18 more.

Does the ThirdMade Shelf MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the ThirdMade Shelf MCP server send data to outside services?

6 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): robots, llms_txt, page_meta, link_graph, scrape and search_verify.

Has the ThirdMade Shelf MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 18, 2026.