agent-exec MCP server
org.448c.agent-exec/agent-execagent-exec.448c.orgstreamable-httpv0.4.0Scanned
Accountless pay-per-use sandbox, storage, locks, attestation, email inbox and DNS names via x402.
28 tools: 1 can delete or cancel things, 3 do something that can't be undone, 23 change things and 1 only read.
- Reads private data · 0
- Reads text others write · 1
- Can send data out · 8
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
1 tool here can delete or cancel things and 3 do something that can't be undone. Govern them.
- 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-exec MCP server can do
The agent-exec MCP server is published by agent-exec.448c.org, a domain the MCP Registry verified. It has 28 tools. It can delete or cancel things (1 tool: storage_replace), do things that cannot be undone, such as sending, paying or publishing (relay_usdc, deadman_register and pay_send) and create or change data (23 tools: service_card, execute, scrape_page, doc_text and 19 more). 1 tool only reads.
1 tool brings in text other people can write (scrape_page); 8 tools can send data out of your system (scrape_page, doc_text, check_url and 5 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, and its tools changed 2 times since Anlyon Radar first saw them.
Last scanned Nov 18, 2026, version 0.4.0.
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.
28 tools
Deletes or cancels 1Claude Code: deny
storage_replacePAID: replace the contents of a stored file you own, keeping its id, with a fresh plan term.No findings
Why it is in this group
- The name "storage_replace" says "replace": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it PAID: replace the contents of a stored file you own, keeping its id, with a fresh plan term. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: from $0.0100 USDC.
Annotations the server declares
- title: Replace a stored object (paid, x402)
- readOnlyHint: false
- destructiveHint: true
- openWorldHint: false
Can't be undone 3Claude Code: ask
deadman_registerPAID: dead-man switch for an irreversible action: register the intent, a lease and a probe; confirm in time or we run the probe and publish the verdict.Can send data outNo findings
Why it is in this group
- The description says "irreversible".
What it can reach
- outboundThe input takes a URL (input.probe.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it PAID: dead-man switch for an irreversible action: register the intent, a lease and a probe; confirm in time or we run the probe and publish the verdict. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0100 USDC.
Annotations the server declares
- title: Register an intent with a verification probe (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
pay_sendPAID: deliver an x402 payment you signed to a seller exactly once, verify it on chain, and get a receipt of what the seller answered.No findings
Why it is in this group
- The name "pay_send" says "pay": a word that usually means money moves, which cannot be taken back.
Description, as the server wrote it PAID: deliver an x402 payment you signed to a seller exactly once, verify it on chain, and get a receipt of what the seller answered. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0200 USDC.
Annotations the server declares
- title: Deliver an x402 authorization you signed, and check the chain (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
relay_usdcPAID: send USDC without gas: you sign an EIP-3009 transfer, we broadcast it on Base with our ETH.Can send data outNo findings
Why it is in this group
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it PAID: send USDC without gas: you sign an EIP-3009 transfer, we broadcast it on Base with our ETH. $0.0200 an order, at most $100. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: Broadcast a USDC transfer you signed, with our gas (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
Changes things 23Claude Code: ask
attest_createPAID: attest a URL: we fetch it and sign what we saw (hash, status, headers, never the body), timestamped.Can send data outNo findings
Why it is in this group
- The name "attest_create" says "create": a word for a change.
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 PAID: attest a URL: we fetch it and sign what we saw (hash, status, headers, never the body), timestamped. A failed fetch is still an observation. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0100 USDC.
Annotations the server declares
- title: Signed observation of a URL (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
check_addressPAID: check a wallet address (EVM) against public risk feeds (OFAC, darklists, on-chain age): verdict, score, reasons.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 PAID: check a wallet address (EVM) against public risk feeds (OFAC, darklists, on-chain age): verdict, score, reasons. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0100 USDC.
Annotations the server declares
- title: Check an EVM address against public risk feeds (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
check_counterpartyPAID: check an x402 quote before you pay it: payee, resource, network, asset and price graded against what you expected.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 PAID: check an x402 quote before you pay it: payee, resource, network, asset and price graded against what you expected. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0100 USDC.
Annotations the server declares
- title: Check a 402 quote against what you expected before you sign it (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
check_urlPAID: check a URL against public risk feeds (phishing lists, domain age, lexical signals): verdict, score, reasons.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 PAID: check a URL against public risk feeds (phishing lists, domain age, lexical signals): verdict, score, reasons. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0100 USDC.
Annotations the server declares
- title: Check a URL against public risk feeds (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
claim_acquirePAID: lock a key for your own fleet (accountless compare-and-set lease).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 PAID: lock a key for your own fleet (accountless compare-and-set lease). outcome "acquired": yours; "held": taken, held.mine says if by you. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: from $0.0100 USDC.
Annotations the server declares
- title: Take a named lock (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
claim_renewPAID: extend a key lock you hold, keeping it.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 PAID: extend a key lock you hold, keeping it. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: from $0.0100 USDC.
Annotations the server declares
- title: Extend a lock you hold (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
create_inboxPAID: disposable email inbox for 24 h, receive only, for verification mail; codes and links are extracted.No findings
Why it is in this group
- The name "create_inbox" says "create": a word for a change.
Description, as the server wrote it PAID: disposable email inbox for 24 h, receive only, for verification mail; codes and links are extracted. Pay again with its id to extend. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0200 USDC.
Annotations the server declares
- title: Get a disposable email address (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
dns_renewPAID: extend a DNS name you own by a year.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 PAID: extend a DNS name you own by a year. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.5000 USDC.
Annotations the server declares
- title: Extend a rented name (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
dns_rentPAID: rent a DNS name, <label>.448c.net, for a year with A/AAAA/CNAME/TXT records; the paying wallet owns it.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 PAID: rent a DNS name, <label>.448c.net, for a year with A/AAAA/CNAME/TXT records; the paying wallet owns it. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.5000 USDC.
Annotations the server declares
- title: Rent a name under 448c.net (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
doc_textPAID: extract text from a PDF, page by page (up to 20 pages), for $0.0100.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
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 PAID: extract text from a PDF, page by page (up to 20 pages), for $0.0100. We fetch it; no OCR. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: Extract the text of one PDF (paid, x402)
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
drift_checkPAID: detect API drift: we re-fetch a venue we snapshot and diff it against our baseline (fields added, removed, moved).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 PAID: detect API drift: we re-fetch a venue we snapshot and diff it against our baseline (fields added, removed, moved). Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0100 USDC.
Annotations the server declares
- title: Re-fetch a venue and diff it against our baseline (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
executePAID: run Python or JavaScript code (python3, node) in an isolated, network-less sandbox; returns stdout, stderr and exit code.No findings
Why it is in this group
- The name "execute" says "execute": a word for a change.
Description, as the server wrote it PAID: run Python or JavaScript code (python3, node) in an isolated, network-less sandbox; returns stdout, stderr and exit code. Price scales with limits. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: Run code (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
handoff_depositPAID: hand off context to a successor process: a typed briefing (done with evidence, not done, in flight, next).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 PAID: hand off context to a successor process: a typed briefing (done with evidence, not done, in flight, next). Anyone with the id reads it; kept 3 days. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0100 USDC.
Annotations the server declares
- title: Deposit a briefing for your successor (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
infer_chatPAID: LLM chat completion for $0.0100, no account or model key: send messages, get {model, content, finish_reason, usage}.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 PAID: LLM chat completion for $0.0100, no account or model key: send messages, get {model, content, finish_reason, usage}. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: One LLM chat completion with no account and no model key (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
listing_promotePAID: promote a directory listing you own above free rows for the plan term; paying again extends.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 PAID: promote a directory listing you own above free rows for the plan term; paying again extends. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: from $0.0667 USDC.
Annotations the server declares
- title: Buy paid placement for a directory listing (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
placement_orderPAID: list your service on agent venues: we submit it to each venue we have a submitter for (17 of 33 today) and prove each entry landed, weekly for 90 days.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 PAID: list your service on agent venues: we submit it to each venue we have a submitter for (17 of 33 today) and prove each entry landed, weekly for 90 days. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.2500 USDC.
Annotations the server declares
- title: Get your service submitted to the agent venues we can reach, and to each new one as we add it (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
relay_permitPAID: approve a USDC spender without gas: you sign an EIP-2612 Permit, we broadcast it on Base.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 PAID: approve a USDC spender without gas: you sign an EIP-2612 Permit, we broadcast it on Base. $0.0200 an approval, at most $100. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: Broadcast a USDC approval you signed, with our gas (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
rpc_callPAID: one blockchain JSON-RPC call on base, base-sepolia, ethereum, solana for $0.0100, no provider account: you name chain and method, we pick the node.No findings
Why it is in this group
- The name "rpc_call" says "call": a word for a change.
Description, as the server wrote it PAID: one blockchain JSON-RPC call on base, base-sepolia, ethereum, solana for $0.0100, no provider account: you name chain and method, we pick the node. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: One JSON-RPC call with no provider account (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
scrape_pagePAID: scrape a web page: readable text, title, links and meta for $0.0100.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 name "scrape_page" 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 PAID: scrape a web page: readable text, title, links and meta for $0.0100. We fetch it (public URLs, up to 2000000 bytes). Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: Read one web page as text (paid, x402)
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
service_cardFREE: what agent-exec sells, each price, and how to pay (x402, USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer, no account).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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it FREE: what agent-exec sells, each price, and how to pay (x402, USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer, no account). Moves no money. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: agent-exec service card
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
storage_buyPAID: store a file (bytes as base64) for the plan term; the paying wallet owns it.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 PAID: store a file (bytes as base64) for the plan term; the paying wallet owns it. Returns its id and expiry. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: from $0.0100 USDC.
Annotations the server declares
- title: Store an object (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
storage_renewPAID: extend a stored file you own by another plan term; same id, contents untouched.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 PAID: extend a stored file you own by another plan term; same id, contents untouched. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: from $0.0100 USDC.
Annotations the server declares
- title: Extend a stored object (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
watch_registerPAID: watch a URL until a condition fires (status, body text, JSON path, any change) or the TTL ends; priced by interval x TTL.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 PAID: watch a URL until a condition fires (status, body text, JSON path, any change) or the TTL ends; priced by interval x TTL. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply. Price: $0.0100 USDC.
Annotations the server declares
- title: Durable poll of a URL until a condition fires (paid, x402)
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
Only reads 1Show 1 read-only tool
flights_lookupPAID: look up a flight: where an aircraft is now (position, altitude, speed, track) by callsign, ICAO24 hex or tail number, $0.0100.No findings
Why it is in this group
- The name "flights_lookup" says "lookup": a word for reading.
Description, as the server wrote it PAID: look up a flight: where an aircraft is now (position, altitude, speed, track) by callsign, ICAO24 hex or tail number, $0.0100. Data: adsb.lol community ADS-B feed, ODbL 1.0. Unpaid, it returns the price; pay with x402 (USDC on Base, Solana, Polygon, Arbitrum, Avalanche, Sei or X Layer) in _meta['x402/payment'] and call again. Feedback: POST /feedback (free). Read daily; no individual reply.
Annotations the server declares
- title: Where an aircraft is right now, by callsign, hex address or tail number (paid, x402)
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
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 write1
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out8
- 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.
Looked at Nov 20, 2026, 19:17 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
- 29
- 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
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 29 tools that change things (1 destructive, 4 irreversible, 24 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://agent-exec.448c.org/mcp: tools/list returned 30 tools to an anonymous client, including storage_replace (destructive), settle_payment (irreversible), relay_usdc (irreversible), deadman_register (irreversible), pay_send (irreversible) and 24 more
Posture changes
- Nov 20, 2026
- Tools that change things, listed without credentials: 41 to 29
- Nov 6, 2026
- Tools that change things, listed without credentials: 38 to 41
- Oct 29, 2026
- Tools that change things, listed without credentials: 37 to 38
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.
PostureRemote posture changed. Tools that change things, listed without credentials: 41 -> 29.View diff
- Tools that change things, listed without credentials: 41 to 29
Tools5 added, 30 removed, 23 changed.+5 added−30 removed~23 changedView diff
- Added:doc_textChanges things
- Added:flights_lookupOnly reads
- Added:infer_chatChanges things
- Added:relay_permitChanges things
- Added:scrape_pageChanges things
- Removed:attest_read
- Removed:claim_acquire_free
- Removed:claim_read
- Removed:claim_release
- Removed:deadman_confirm
- Removed:deadman_read
- Removed:dns_lookup
- Removed:dns_release
- Removed:dns_rent_free
- Removed:dns_set_records
- Removed:drift_baseline
- Removed:drift_venues
- Removed:handoff_read
- Removed:inbox_status
- Removed:pay_receipt
- Removed:pay_requirements
- Removed:placement_cancel
- Removed:placement_ready
- Removed:placement_status
- Removed:placement_venues
- Removed:read_inbox
- Removed:release_inbox
- Removed:search_listings
- Removed:storage_delete
- Removed:storage_list
- Removed:storage_read
- Removed:storage_usage
- Removed:watch_read
- Removed:x402_supported
- Removed:x402_verify
- Changed:attest_createChanges thingsdescriptioninputSchema
- Changed:check_addressChanges thingsdescriptioninputSchema
- Changed:check_counterpartyChanges thingsdescriptioninputSchema
- Changed:check_urlChanges thingsdescriptioninputSchema
- Changed:claim_acquireChanges thingsdescriptioninputSchema
- Changed:claim_renewChanges thingsdescriptioninputSchema
- Changed:create_inboxChanges thingsdescription
- Changed:deadman_registerCan't be undonedescriptioninputSchema
- Changed:dns_renewChanges thingsdescriptioninputSchema
- Changed:dns_rentChanges thingsdescriptioninputSchema
- Changed:drift_checkChanges thingsdescriptioninputSchema
- Changed:executeChanges thingsdescriptioninputSchema
- Changed:handoff_depositChanges thingsdescriptioninputSchema
- Changed:listing_promoteChanges thingsdescriptioninputSchema
- Changed:pay_sendCan't be undonedescriptioninputSchema
- Changed:placement_orderChanges thingsdescriptioninputSchema
- Changed:relay_usdcCan't be undonedescriptioninputSchema
- Changed:rpc_callChanges thingsdescriptioninputSchema
- Changed:service_cardChanges thingsdescription
- Changed:storage_buyChanges thingsdescriptioninputSchema
- Changed:storage_renewChanges thingsdescriptioninputSchema
- Changed:storage_replaceDeletes or cancelsdescriptioninputSchema
- Changed:watch_registerChanges thingsdescriptioninputSchema
PostureRemote posture changed. Tools that change things, listed without credentials: 38 -> 41.View diff
- Tools that change things, listed without credentials: 38 to 41
Tools5 added, 1 removed, 3 changed.+5 added−1 removed~3 changedView diff
- Added:check_counterpartyChanges things
- Added:relay_usdcCan't be undone
- Added:rpc_callChanges things
- Added:x402_supportedChanges things
- Added:x402_verifyChanges things
- Removed:free_inbox
- Changed:placement_orderChanges thingsdescription
- Changed:search_listingsOnly readsdescription
- Changed:service_cardChanges thingsdescription
PostureRemote posture changed. Tools that change things, listed without credentials: 37 -> 38.View diff
- Tools that change things, listed without credentials: 37 to 38
Install
Pinned to version 0.4.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http agent-exec https://agent-exec.448c.org/mcpclaude mcp add --transport http agent-exec https://agent-exec.448c.org/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-exec__flights_lookup"
],
"ask": [
"mcp__agent-exec__service_card",
"mcp__agent-exec__execute",
"mcp__agent-exec__scrape_page",
"mcp__agent-exec__doc_text",
"mcp__agent-exec__infer_chat",
"mcp__agent-exec__storage_buy",
"mcp__agent-exec__storage_renew",
"mcp__agent-exec__create_inbox",
"mcp__agent-exec__dns_rent",
"mcp__agent-exec__dns_renew",
"mcp__agent-exec__check_address",
"mcp__agent-exec__check_url",
"mcp__agent-exec__check_counterparty",
"mcp__agent-exec__rpc_call",
"mcp__agent-exec__relay_usdc",
"mcp__agent-exec__relay_permit",
"mcp__agent-exec__listing_promote",
"mcp__agent-exec__drift_check",
"mcp__agent-exec__watch_register",
"mcp__agent-exec__claim_acquire",
"mcp__agent-exec__claim_renew",
"mcp__agent-exec__attest_create",
"mcp__agent-exec__handoff_deposit",
"mcp__agent-exec__deadman_register",
"mcp__agent-exec__pay_send",
"mcp__agent-exec__placement_order"
],
"deny": [
"mcp__agent-exec__storage_replace"
]
}
}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-exec. Review it before you use it.
Capability tags behind 8 rules
- mcp__agent-exec__scrape_page untrusted-input, outbound
- mcp__agent-exec__doc_text outbound
- mcp__agent-exec__check_url outbound
- mcp__agent-exec__relay_usdc outbound
- mcp__agent-exec__watch_register outbound
- mcp__agent-exec__attest_create outbound
- mcp__agent-exec__deadman_register outbound
- mcp__agent-exec__placement_order 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/org.448c.agent-exec/agent-exec/verify?hash=1be3ae40180f66b5d86dcb96014ae3b93e97260f6b6423f98aca83d0a2e1b00b" | jq -e '.changed == false'Pinned hash 1be3ae40180f, from Nov 18, 2026, 23:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/org.448c.agent-exec/agent-exec/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/org.448c.agent-exec/agent-exec)Source: MCP Registry. Synced Oct 21, 2026, 18:54 UTC.
agent-exec MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the agent-exec MCP server safe?
- No scan can promise that. What its definitions show: 28 tools, 27 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 agent-exec MCP server delete data?
- Yes. 1 tool can delete, cancel or revoke something: storage_replace.
Can the agent-exec MCP server do things that cannot be undone?
- Yes. 3 tools can send, pay, publish or merge: relay_usdc, deadman_register and pay_send.
What tools does the agent-exec MCP server have?
- 28 tools: 1 destructive, 3 irreversible, 23 write, 1 read. storage_replace, relay_usdc, deadman_register, pay_send, service_card, execute, scrape_page, doc_text and 20 more.
Does the agent-exec MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the agent-exec MCP server send data to outside services?
- 8 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): scrape_page, doc_text, check_url, relay_usdc, watch_register, attest_create and 2 more.
Has the agent-exec MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 18, 2026. Every change is on this page and in its Atom feed.