Agent Data Tools MCP server
io.github.Nutshell2000/agent-data-toolsNutshell2000streamable-httpv1.331.0Scanned
673 agent tools: chain data, prices, sanctions, web, PDF. 10 free calls, then from $0.001.
15 tools: 1 do something that can't be undone, 8 change things and 6 only read.
- Reads private data · 1
- Reads text others write · 13
- Can send data out · 4
- 5 findings
Tools listed without sign-in. Remote posture
With Anlyon
1 tool here does something that can't be undone. Govern it.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
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 Agent Data Tools MCP server can do
The Agent Data Tools MCP server is published by the GitHub account Nutshell2000. It has 15 tools. It can do things that cannot be undone, such as sending, paying or publishing (ai_chat) and create or change data (8 tools: call_tool, news, weather, price_token and 4 more), but cannot delete data. 6 tools only read.
1 tool reads private data (extract); 13 tools bring in text other people can write (call_tool, search, extract and 10 more); 4 tools can send data out of your system (extract, transcribe, pdf_text and 1 more). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.
The fixed checks matched 5 findings (tool-shadowing, annotation-conflict and base64-blob). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed 10 times since Anlyon Radar first saw them.
Last scanned Nov 21, 2026, version 1.331.0.
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.
15 tools
Can't be undone 1Claude Code: ask
ai_chatSend a prompt, get an answer from Meta's Llama 4 Scout (17B active, 16 experts), with an optional system prompt, up to 1,024 output tokens and adjustable temperature.Reads text others writeCan send data out1 finding
Why it is in this group
- The description's first verb is "send": a word for sending, and what is sent cannot be taken back.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe description's first verb is "send": it sends or publishes something to other people.
Findings
- Warn
The server marks "ai_chat" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The description's first verb is "send": a word for sending, and what is sent cannot be taken back.
Description, as the server wrote it Send a prompt, get an answer from Meta's Llama 4 Scout (17B active, 16 experts), with an optional system prompt, up to 1,024 output tokens and adjustable temperature. Fast and cheap. Use for drafting, reasoning steps, Q&A and any quick LLM call without an AI provider account; other open models by name at /v1/chat/completions, harder reasoning at ai_chat_large. Example: prompt=Explain HTTP status code 402 in one sentence.. Price $0.002/call via x402 USDC or a card key (/credits); failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Chat completions (LLM): Llama 4 Scout
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
Changes things 8Claude Code: ask
call_toolRun one of this service's tools by name.Reads text others write2 findings
Why it is in this group
- The name "call_tool" says "call": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The input schema at properties.max_price.description contains "every other tool", wording about other tools or servers.
What this check looks forMatched text
inputSchema.properties.max_price.description: "..., for example "$1.00". Leave it out for every other tool"
- Warn
The server marks "call_tool" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The name "call_tool" says "call": a word for a change.
Description, as the server wrote it Run one of this service's tools by name. Costs that tool's own price per successful call (most $0.01; describe_tool and find_tool show it); a failed call costs nothing, and tools marked inFreeCalls (tools that only compute or look up public data; not AI, speech, transcription, images, rendering, search or resold data) run free while the caller has free calls left. Pass the tool's inputs as "arguments". If you do not know the name or the inputs, call find_tool or describe_tool first: a wrong name or a missing input is refused before anything is charged. A tool that costs more than $0.10 a call (find_tool marks it needsMaxPrice) runs only when you pass that price as "max_price"; without it the call is refused, uncharged, and the refusal states the price.
Annotations the server declares
- title: Run a tool by name
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
- idempotentHint: false
check_payeeFast check of a payee before an agent pays it: OFAC SDN match on the address, account type (wallet, delegated wallet or contract) and nonce on Base, Ethereum, Arbitrum, Optimism or…Reads text others write1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The description contains a 42-character base64 string.
What this check looks forMatched text
description: "...y. Example: address=0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045. Price $0.005/call..."
Description, as the server wrote it Fast check of a payee before an agent pays it: OFAC SDN match on the address, account type (wallet, delegated wallet or contract) and nonce on Base, Ethereum, Arbitrum, Optimism or Polygon, and optionally the payee domain's age and trust flags. One block, review or pass verdict with reasons. Screening aid, not legal advice. Use before an x402 or USDC payment to an unfamiliar address; holdings and wallet heuristics: check_counterparty. Example: address=0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045. Price $0.005/call via x402 USDC or a card key (/credits), first 10 free; failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Payee screening: OFAC-screen an address before paying, plus account type and domain age
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
newsNews headlines on any topic, crypto included, as JSON: title, URL, snippet, publisher domain and age, from the Brave Search API's news index.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it News headlines on any topic, crypto included, as JSON: title, URL, snippet, publisher domain and age, from the Brave Search API's news index. 1-20 results per call; freshness limits results to the past day, week or month. Results are for the caller's immediate use only: no storing, resale or model training (Brave API terms). Upstream failures return 503, uncharged. Use when an agent needs current news on a topic. Example: q=bitcoin ETF. Price $0.01/call via x402 USDC or a card key (/credits); failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: News headlines, latest news: search current news articles (Brave)
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
pdf_textExtract the text of a public PDF (up to 10 MB) from its URL, page by page, with title, author and page count.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- 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 Extract the text of a public PDF (up to 10 MB) from its URL, page by page, with title, author and page count. Choose the first page, how many pages (up to 200) and maxChars. arXiv abstract links work. If a site blocks our fetch, the Internet Archive's dated copy is read and the answer says so. Scanned, encrypted and non-PDF files fail uncharged. Use to read papers, invoices, reports and manuals into an agent. Example: url=https://bitcoin.org/bitcoin.pdf. Price $0.005/call via x402 USDC or a card key (/credits), first 10 free; failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: PDF to text, PDF text extraction: extract the text of a PDF document from a URL, page by page
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
price_tokenToken price in USD for a crypto token by symbol, coin name or CoinGecko id from a built-in list of main tokens (ETH, BTC, USDC, AERO, ARB, OP, LINK, PEPE…) or any ERC-20 0x address…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Token price in USD for a crypto token by symbol, coin name or CoinGecko id from a built-in list of main tokens (ETH, BTC, USDC, AERO, ARB, OP, LINK, PEPE…) or any ERC-20 0x address on Base, Ethereum, Arbitrum or Optimism, from the deepest Uniswap v3 pool against USDC or WETH, read live. Returns pool, fee tier, reserve and a thin-liquidity flag. Spot price; no pool is 422, uncharged. Use to value a token or check a quote before a swap or payment. Example: symbol=ETH. Price $0.002/call via x402 USDC or a card key (/credits), first 10 free; failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Crypto price and token price: USD price of a token from Uniswap v3 pools
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
sanctions_checkCheck a crypto wallet against the OFAC SDN list before you pay it (EVM, BTC, TRX), or screen a person or company name, with aliases, against OFAC, the EU consolidated list, the UN …Reads text others write1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The description contains a 42-character base64 string.
What this check looks forMatched text
description: "...y. Example: address=0x098B716B8Aaf21512996dC57EB0615e2383E2f96. Price $0.009/call..."
Description, as the server wrote it Check a crypto wallet against the OFAC SDN list before you pay it (EVM, BTC, TRX), or screen a person or company name, with aliases, against OFAC, the EU consolidated list, the UN Security Council list and the UK Sanctions List in one call. Returns a verdict (block, review, pass), each match with list, list ID, programme and strength, and every list's date. Screening aid, not legal advice. Use before paying; with wallet risk: check_counterparty. Example: address=0x098B716B8Aaf21512996dC57EB0615e2383E2f96. Price $0.009/call via x402 USDC or a card key (/credits), first 10 free; failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Sanctions check (OFAC screening): SDN screening for a crypto wallet or name
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
transcribeTranscribe speech from an audio file at a public URL, or uploaded in a POST body, using Whisper large-v3-turbo.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- 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 Transcribe speech from an audio file at a public URL, or uploaded in a POST body, using Whisper large-v3-turbo. Returns the transcript text, detected language, duration and word count; subtitles=srt or vtt adds subtitle text and timestamps=true adds timed segments and words. Accepts common formats such as MP3, WAV, M4A and OGG up to 5 MB, in 90+ languages. Use for voice notes, podcast clips, meeting snippets, subtitles and voice messages. Example: url=https://commons.wikimedia.org/wiki/Special:FilePath/En-us-hello.ogg. Price $0.01/call via x402 USDC or a card key (/credits); failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Speech to text, transcribe audio: Whisper transcription of an audio file
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
weatherCurrent weather by city name, or by latitude and longitude, for any place on Earth: temperature in Celsius, conditions (clear, cloudy, rain, snow), precipitation in mm, wind, humid…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Current weather by city name, or by latitude and longitude, for any place on Earth: temperature in Celsius, conditions (clear, cloudy, rain, snow), precipitation in mm, wind, humidity, cloud cover and air pressure, plus an hourly forecast for up to 48 hours. Data from the Norwegian Meteorological Institute; city names resolve via Wikidata. Use for travel and event planning, logistics, agriculture and any agent that needs the weather. Example: city=London. Price $0.002/call via x402 USDC or a card key (/credits), first 10 free; failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Weather forecast: current weather and 48-hour hourly forecast for any city or coordinates
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
Only reads 6Show 6 read-only tools
company_searchSearch a company name across official registers in one call: France, Sweden, Norway, Finland, Czechia and Estonia, plus GLEIF LEI records for every other country.Reads text others writeNo findings
Why it is in this group
- The name "company_search" says "search": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Search a company name across official registers in one call: France, Sweden, Norway, Finland, Czechia and Estonia, plus GLEIF LEI records for every other country. One ranked list: register ID, legal form, city, LEI, a link to the register's own page, time read. A source that does not answer is named; the rest still answer. Use for KYB, supplier and customer checks, and facts about a European company. Example: q=Spotify. Price $0.01/call via x402 USDC or a card key (/credits), first 10 free; failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Company search, EU official registers
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
describe_toolEverything needed to call one tool: what it does, every input with its type, whether it is required and an example value, the price per successful call (and the lower price with a …No findings
Why it is in this group
- The name "describe_tool" says "describe": a word for reading.
Description, as the server wrote it Everything needed to call one tool: what it does, every input with its type, whether it is required and an example value, the price per successful call (and the lower price with a prepaid key where there is one), whether it is in the free calls (they run tools that only compute or look up public data), an example call and the shape of its answer. Free. Take the name from find_tool.
Annotations the server declares
- title: Inputs, price and example of one tool
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
extractFetch any public web page URL and get clean, LLM-ready text or Markdown: main content, code blocks, tables, links and headings, for RAG and agent context.Reads private dataReads 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
- private-dataThe description's first sentence says "fetch" and "table": it reads databases.
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe description's first sentence says "fetch" and "web": 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 Fetch any public web page URL and get clean, LLM-ready text or Markdown: main content, code blocks, tables, links and headings, for RAG and agent context. Static HTML, Markdown or text; optional section ID. Default 20,000 output characters, max 100,000; input 2 MiB. No JavaScript rendering (see /extract/rendered), logins or paywalls. Empty pages, bad sections, browser challenges and errors are not charged. Use to read docs or articles. Example: url=https://example.com. Price $0.002/call via x402 USDC or a card key (/credits), first 10 free; failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Extract web content, Markdown: any URL to clean text for LLMs
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
find_toolFind which of this service's 673 tools does a job.No findings
Why it is in this group
- The name "find_tool" says "find": a word for reading.
Description, as the server wrote it Find which of this service's 673 tools does a job. Say the job in plain words ("current weather in Paris", "text of a PDF", "is this wallet sanctioned") and get up to five tools by default (pass count for more, up to ten), best match first, each with its name for call_tool, what it does, its price and its required inputs. Free; call it before call_tool whenever you do not already know the tool's name.
Annotations the server declares
- title: Find a tool for a job
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
searchExa-style web search: top 1-20 results with title, URL, snippet, published date and favicon (Brave index); optional domain, date, category and phrase filters; optional live page te…Reads text others writeNo findings
Why it is in this group
- The name "search" says "search": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe description's first sentence says "web": it reads web pages, which anyone can write.
Description, as the server wrote it Exa-style web search: top 1-20 results with title, URL, snippet, published date and favicon (Brave index); optional domain, date, category and phrase filters; optional live page text, highlights and AI summaries of the top 10 pages. Results are for immediate use only: no storing, resale or model training (Brave API terms). Failures return 503, uncharged. Use when an agent needs current information or sources from the open web. Example: q=x402 payment protocol. Price $0.01/call via x402 USDC or a card key (/credits); failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: Web search API with contents: Exa-style search, filters, page text and highlights
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
vat_validateCheck an EU VAT number against the European Commission's VIES service: valid or not, with the registered name and address where the member state shares them.Reads text others writeNo findings
Why it is in this group
- The name "vat_validate" says "validate": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Check an EU VAT number against the European Commission's VIES service: valid or not, with the registered name and address where the member state shares them. Any spacing (SE556703748501, DE 123 456 789; EL or GR for Greece, XI for Northern Ireland). Or verify a business by name: found or not in the French, Swedish, Norwegian and Finnish registers. Use to check a B2B customer before invoicing without VAT, or to confirm a supplier. Example: vat=SE556703748501. Price $0.01/call via x402 USDC or a card key (/credits), first 10 free; failed calls free. Reads or computes only and changes nothing; each call is charged, so repeats are not free.
Annotations the server declares
- title: VAT number validation, EU VIES
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
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.
52 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 private-data tool breaks every one.
- Reads private data1
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write13
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out4
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__agent-data-tools__extract"
]
}
}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.
Looked at Nov 20, 2026, 18:59 UTC, unchanged since Nov 20, 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
- 9
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://agent-data-tools.revmesh2074.workers.dev/.well-known/oauth-protected-resource
- Authorization servers
- https://agent-data-tools.revmesh2074.workers.dev
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- yes
- Client ID metadata documents
- no
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 9 tools that change things (1 irreversible, 8 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://agent-data-tools.revmesh2074.workers.dev/mcp: tools/list returned 15 tools to an anonymous client, including ai_chat (irreversible), call_tool (write), news (write), weather (write), price_token (write) and 4 more
Posture changes
- Nov 20, 2026
- Tools that change things, listed without credentials: 380 to 9
- Nov 13, 2026
- Tools that change things, listed without credentials: 299 to 380
- Nov 6, 2026
- Tools that change things, listed without credentials: 268 to 299
- Oct 29, 2026
- Tools that change things, listed without credentials: 145 to 268
Findings
Matches of fixed checks, with the text they matched.
- Warncall_tool
The input schema at properties.max_price.description contains "every other tool", wording about other tools or servers.
What this check looks forMatched text
inputSchema.properties.max_price.description: "..., for example "$1.00". Leave it out for every other tool"
- Warncall_tool
The server marks "call_tool" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The name "call_tool" says "call": a word for a change.
- Warnsanctions_check
The description contains a 42-character base64 string.
What this check looks forMatched text
description: "...y. Example: address=0x098B716B8Aaf21512996dC57EB0615e2383E2f96. Price $0.009/call..."
- Warncheck_payee
The description contains a 42-character base64 string.
What this check looks forMatched text
description: "...y. Example: address=0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045. Price $0.005/call..."
- Warnai_chat
The server marks "ai_chat" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The description's first verb is "send": a word for sending, and what is sent cannot be taken back.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools11 changed.~11 changedView diff
- Changed:ai_chatCan't be undonedescription
- Changed:call_toolChanges thingsdescription
- Changed:check_payeeChanges thingsdescription
- Changed:company_searchOnly readsdescription
- Changed:describe_toolOnly readsdescription
- Changed:extractOnly readsdescription
- Changed:pdf_textChanges thingsdescription
- Changed:price_tokenChanges thingsdescription
- Changed:sanctions_checkChanges thingsdescription
- Changed:vat_validateOnly readsdescription
- Changed:weatherChanges thingsdescription
PostureRemote posture changed. Tools that change things, listed without credentials: 380 -> 9.View diff
- Tools that change things, listed without credentials: 380 to 9
Tools1 changed.~1 changedView diff
- Changed:find_toolOnly readsdescription
Tools4 changed.~4 changedView diff
- Changed:check_payeeChanges thingsdescription
- Changed:pdf_textChanges thingsdescription
- Changed:price_tokenChanges thingsdescription
- Changed:sanctions_checkChanges thingsdescription
Tools5 changed.~5 changedView diff
- Changed:check_payeeChanges thingsdescription
- Changed:find_toolOnly readsdescription
- Changed:pdf_textChanges thingsdescription
- Changed:price_tokenChanges thingsdescription
- Changed:sanctions_checkChanges thingsdescription
Tools1 changed.~1 changedView diff
- Changed:find_toolOnly readsdescription
Tools3 changed.~3 changedView diff
- Changed:find_toolOnly readsdescription
- Changed:pdf_textChanges thingsinputSchema
- Changed:price_tokenChanges thingsinputSchema
Tools1 changed.~1 changedView diff
- Changed:find_toolOnly readsdescription
Tools1 changed.~1 changedView diff
- Changed:find_toolOnly readsdescription
Install
Pinned to version 1.331.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http agent-data-tools https://agent-data-tools.revmesh2074.workers.dev/mcpclaude mcp add --transport http agent-data-tools https://agent-data-tools.revmesh2074.workers.dev/mcpRun 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.
{
"permissions": {
"allow": [
"mcp__agent-data-tools__find_tool",
"mcp__agent-data-tools__describe_tool",
"mcp__agent-data-tools__search",
"mcp__agent-data-tools__company_search",
"mcp__agent-data-tools__vat_validate"
],
"ask": [
"mcp__agent-data-tools__call_tool",
"mcp__agent-data-tools__extract",
"mcp__agent-data-tools__news",
"mcp__agent-data-tools__weather",
"mcp__agent-data-tools__price_token",
"mcp__agent-data-tools__sanctions_check",
"mcp__agent-data-tools__check_payee",
"mcp__agent-data-tools__transcribe",
"mcp__agent-data-tools__pdf_text",
"mcp__agent-data-tools__ai_chat"
],
"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 agent-data-tools. Review it before you use it.
Capability tags behind 13 rules
- mcp__agent-data-tools__call_tool untrusted-input
- mcp__agent-data-tools__search untrusted-input
- mcp__agent-data-tools__extract private-data, untrusted-input, outbound
- mcp__agent-data-tools__news untrusted-input
- mcp__agent-data-tools__weather untrusted-input
- mcp__agent-data-tools__price_token untrusted-input
- mcp__agent-data-tools__sanctions_check untrusted-input
- mcp__agent-data-tools__check_payee untrusted-input
- mcp__agent-data-tools__transcribe untrusted-input, outbound
- mcp__agent-data-tools__pdf_text untrusted-input, outbound
- mcp__agent-data-tools__company_search untrusted-input
- mcp__agent-data-tools__vat_validate untrusted-input
- mcp__agent-data-tools__ai_chat 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/io.github.Nutshell2000/agent-data-tools/verify?hash=b85658f5fee049a7f9f5e1191fc57442c36960363dfe4f870a8a0d1a4405809f" | jq -e '.changed == false'Pinned hash b85658f5fee0, from Nov 21, 2026, 19:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.Nutshell2000/agent-data-tools/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.Nutshell2000/agent-data-tools)Source: MCP Registry · repository. Synced Nov 21, 2026, 19:17 UTC.
Agent Data Tools MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Agent Data Tools MCP server safe?
- No scan can promise that. What its definitions show: 15 tools, 9 of which can change, send or delete something, and 5 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 Agent Data Tools MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Agent Data Tools MCP server do things that cannot be undone?
- Yes. 1 tool can send, pay, publish or merge: ai_chat.
What tools does the Agent Data Tools MCP server have?
- 15 tools: 1 irreversible, 8 write, 6 read. ai_chat, call_tool, news, weather, price_token, sanctions_check, check_payee, transcribe and 7 more.
Does the Agent Data Tools MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Agent Data Tools MCP server send data to outside services?
- 4 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): extract, transcribe, pdf_text and ai_chat.
Has the Agent Data Tools MCP server changed its tools?
- Yes, 10 times since Anlyon Radar started tracking it, most recently on Nov 21, 2026. Every change is on this page and in its Atom feed.