Skip to content

AgentWatch MCP server

dev.workers.agentwatch.agentwatch/agentwatchagentwatch.agentwatch.workers.devstreamable-httpv0.5.1Scanned

Read-only watchtower for AI agents on-chain: decoded receipts, plan vs execution, Safe audits.

21 tools: 1 can delete or cancel things, 7 change things and 13 only read.

Tools: 1 delete, 0 irreversible, 7 write, 13 read.
  • Reads private data · 0
  • Reads text others write · 0
  • Can send data out · 0
  • No findings

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here can delete or cancel things. Govern it.

  • 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 AgentWatch MCP server can do

The AgentWatch MCP server is published by agentwatch.agentwatch.workers.dev, a domain the MCP Registry verified. It has 21 tools. It can delete or cancel things (1 tool: remove_recovery_owner) and create or change data (7 tools: run_audit, register_intent, add_watch, watch_demo_set and 3 more). 13 tools only read.

None of its tools reads private data, brings in text other people write, or sends data out.

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 15, 2026, version 0.5.1.

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.

21 tools

Deletes or cancels 1Claude Code: deny
  • remove_recovery_ownerRemove a saved recovery address for this user.No findings

    Why it is in this group

    • The name "remove_recovery_owner" says "remove": a word for removing or ending something.
    Description, as the server wrote it
    Remove a saved recovery address for this user.
Changes things 7Claude Code: ask
  • add_watchStart watching an address.No findings

    Why it is in this group

    • The name "add_watch" says "add": a word for a change.
    Description, as the server wrote it
    Start watching an address. Use when onboarding a new agent or Safe.
  • refresh_watchesForce an immediate poller pass over this principal's watched addresses (catch up txs from explorers into AgentWatch).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
    Force an immediate poller pass over this principal's watched addresses (catch up txs from explorers into AgentWatch). Use after add_watch / a fresh broadcast when get_feed is still empty. Optional address / chains to focus the pass.
  • register_intentDeclare an intent BEFORE broadcasting.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
    Declare an intent BEFORE broadcasting. Returns {intent_id, pairing_code}. Append the 16-hex pairing_code as the last 8 bytes of calldata (no magic prefix), then broadcast — EXCEPT bare ETH transfers to contracts (empty data + value>0): appending reverts on Safe/fallback handlers; register without a suffix and use expected.steps for multi-leg flows (e.g. approve+supply). Pairing codes are single-use and only consumed by successful (non-reverted) txs. Use deadline_minutes (relative) — normalized to absolute constraints.deadline at registration.
  • run_auditRun provenance audit on a Safe/address.No findings

    Why it is in this group

    • The name "run_audit" says "run": a word for a change.
    Description, as the server wrote it
    Run provenance audit on a Safe/address. Use when the user asks 'is this fake?' or wants a shareable verdict.
  • set_agent_passportDeclare who a watched agent is: model provider, run cadence, role, purpose, operator, and whether to publish a shareable public passport at /a/<slug>.No findings

    Why it is in this group

    • The name "set_agent_passport" says "set": a word for a change.
    Description, as the server wrote it
    Declare who a watched agent is: model provider, run cadence, role, purpose, operator, and whether to publish a shareable public passport at /a/<slug>. Declarations never change the receipt-derived numbers next to them; publishing exposes only this address, never the rest of the watch list.
  • set_recovery_ownerSave or update a recovery address for this user.No findings

    Why it is in this group

    • The name "set_recovery_owner" says "set": a word for a change.
    Description, as the server wrote it
    Save or update a recovery address for this user. Product default Safe is then 1/2: threshold 1, owners = [agent, this recovery]. Use make_default=true to prefer this label in get_safe_defaults.
  • watch_demo_setONE-SHOT cold-start for brand-new Claude / MCP users with an empty watch list.No findings

    Why it is in this group

    • The name "watch_demo_set" says "set": a word for a change.
    Description, as the server wrote it
    ONE-SHOT cold-start for brand-new Claude / MCP users with an empty watch list. Adds the July 18 fixture (Agent EOA + Safe A + Safe B on Base, with declared intents) and one live Polygon trading bot, so the first dashboard has recent activity. Call this FIRST after connect when list_watched is empty, then call get_dashboard (Always allow the App). Idempotent — skips addresses already watched.
Only reads 13Show 13 read-only tools
  • get_agent_passportRead the declared identity of a watched agent (name, model provider, how often it runs, role, purpose, operator) and whether its public passport page is published.No findings

    Why it is in this group

    • The name "get_agent_passport" says "get": a word for reading.
    Description, as the server wrote it
    Read the declared identity of a watched agent (name, model provider, how often it runs, role, purpose, operator) and whether its public passport page is published.
  • get_alertsList alerts scoped to this principal's watched addresses (empty watch list → call watch_demo_set first).No findings

    Why it is in this group

    • The name "get_alerts" says "get": a word for reading.
    Description, as the server wrote it
    List alerts scoped to this principal's watched addresses (empty watch list → call watch_demo_set first). Use when checking what needs attention.
  • get_benchmarkCounterfactual P&L / agent alpha vs do-nothing baselines.No findings

    Why it is in this group

    • The name "get_benchmark" says "get": a word for reading.
    Description, as the server wrote it
    Counterfactual P&L / agent alpha vs do-nothing baselines. Use for performance honesty checks. Check matched_agent — false means the key has no indexed activity.
  • get_capabilitiesBlast-radius / historical capability for an agent key.No findings

    Why it is in this group

    • The name "get_capabilities" says "get": a word for reading.
    Description, as the server wrote it
    Blast-radius / historical capability for an agent key. Use before trusting an agent with funds.
  • get_dashboardPersona-tailored dashboard brief (fidelity, alerts, pending plans, recent planned-vs-executed).No findings

    Why it is in this group

    • The name "get_dashboard" says "get": a word for reading.
    Description, as the server wrote it
    Persona-tailored dashboard brief (fidelity, alerts, pending plans, recent planned-vs-executed). NEW USERS with empty watches: call watch_demo_set first, then get_dashboard again. Narrate in plain language; ask the user to Always allow the AgentWatch MCP App widget when the host prompts.
  • get_feedDecoded activity feed for an address.No findings

    Why it is in this group

    • The name "get_feed" says "get": a word for reading.
    Description, as the server wrote it
    Decoded activity feed for an address. Use when reviewing what an agent did on-chain.
  • get_fidelityIntent↔execution fidelity score for an agent or Safe.No findings

    Why it is in this group

    • The name "get_fidelity" says "get": a word for reading.
    Description, as the server wrote it
    Intent↔execution fidelity score for an agent or Safe. Use for accountability / coverage meter. Check matched_agent — false means the address has no indexed activity (as signer or as the Safe that executed), so the score is vacuous.
  • get_leaderboardPublic standings for agents with a published passport, ranked from receipts (declared-first coverage, fidelity, reliability, hygiene, sustained work).No findings

    Why it is in this group

    • The name "get_leaderboard" says "get": a word for reading.
    Description, as the server wrote it
    Public standings for agents with a published passport, ranked from receipts (declared-first coverage, fidelity, reliability, hygiene, sustained work). Pass agent_key to get that agent's own rank, its component scores, and what it would need to climb. Read-only: nothing here can be self-reported.
  • get_pending_proposalsPending Safe multisig proposals from the Safe Transaction Service (indexer confidence).No findings

    Why it is in this group

    • The name "get_pending_proposals" says "get": a word for reading.
    Description, as the server wrote it
    Pending Safe multisig proposals from the Safe Transaction Service (indexer confidence). Pass safe=0x… or omit to scan all watched addresses.
  • get_safe_defaultsReturn the default Safe ownership plan for an agent EOA: threshold 1 + agent + selected recovery (1/2 when one recovery is set).No findings

    Why it is in this group

    • The name "get_safe_defaults" says "get": a word for reading.
    Description, as the server wrote it
    Return the default Safe ownership plan for an agent EOA: threshold 1 + agent + selected recovery (1/2 when one recovery is set). Call before register_intent for Safe deploys. AgentWatch never deploys or signs.
  • get_tcaTCA market-context snapshot for an event_id (chain:txHash).No findings

    Why it is in this group

    • The name "get_tca" says "get": a word for reading.
    Description, as the server wrote it
    TCA market-context snapshot for an event_id (chain:txHash). Use when judging execution quality.
  • list_recovery_ownersList this user's saved recovery wallets (human co-signers for default 1/2 Safes).No findings

    Why it is in this group

    • The name "list_recovery_owners" says "list": a word for reading.
    Description, as the server wrote it
    List this user's saved recovery wallets (human co-signers for default 1/2 Safes). Each may have a label/purpose — users often keep several for different vaults.
  • list_watchedList watched addresses for this connector principal.No findings

    Why it is in this group

    • The name "list_watched" says "list": a word for reading.
    Description, as the server wrote it
    List watched addresses for this connector principal. NEW USERS: if empty, call watch_demo_set next (do not invent a global feed), then get_dashboard.

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 or untrusted-input or outbound.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
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, 18:53 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
8
Protocol version
2025-03-26
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
https://agentwatch.agentwatch.workers.dev/.well-known/oauth-protected-resource/mcp
Authorization servers
https://agentwatch.agentwatch.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 8 tools that change things (1 destructive, 7 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://agentwatch.agentwatch.workers.dev/mcp: tools/list returned 21 tools to an anonymous client, including remove_recovery_owner (destructive), run_audit (write), register_intent (write), add_watch (write), watch_demo_set (write) and 3 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 0.5.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http agentwatch https://agentwatch.agentwatch.workers.dev/mcp
claude mcp add --transport http agentwatch https://agentwatch.agentwatch.workers.dev/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__agentwatch__list_watched",
      "mcp__agentwatch__get_feed",
      "mcp__agentwatch__get_capabilities",
      "mcp__agentwatch__get_alerts",
      "mcp__agentwatch__get_pending_proposals",
      "mcp__agentwatch__get_fidelity",
      "mcp__agentwatch__get_leaderboard",
      "mcp__agentwatch__get_benchmark",
      "mcp__agentwatch__get_tca",
      "mcp__agentwatch__get_agent_passport",
      "mcp__agentwatch__list_recovery_owners",
      "mcp__agentwatch__get_safe_defaults",
      "mcp__agentwatch__get_dashboard"
    ],
    "ask": [
      "mcp__agentwatch__run_audit",
      "mcp__agentwatch__register_intent",
      "mcp__agentwatch__add_watch",
      "mcp__agentwatch__watch_demo_set",
      "mcp__agentwatch__refresh_watches",
      "mcp__agentwatch__set_agent_passport",
      "mcp__agentwatch__set_recovery_owner"
    ],
    "deny": [
      "mcp__agentwatch__remove_recovery_owner"
    ]
  }
}

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 agentwatch. Review it before you use it.

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/dev.workers.agentwatch.agentwatch/agentwatch/verify?hash=c8356af0888aac17277f4ce1822bfbb284c14c5be688a024bccfda3307619446" | jq -e '.changed == false'

Pinned hash c8356af0888a, from Nov 15, 2026, 04:18 UTC.

Watch

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

https://radar.anlyon.com/servers/dev.workers.agentwatch.agentwatch/agentwatch/changes.xml

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

Badge

Anlyon Radar badge for dev.workers.agentwatch.agentwatch/agentwatch
[![Anlyon Radar](https://radar.anlyon.com/badge/dev.workers.agentwatch.agentwatch/agentwatch.svg)](https://radar.anlyon.com/servers/dev.workers.agentwatch.agentwatch/agentwatch)

Source: MCP Registry. Synced Oct 21, 2026, 18:12 UTC.

AgentWatch MCP server: questions and answers

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

Is the AgentWatch MCP server safe?

No scan can promise that. What its definitions show: 21 tools, 8 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 AgentWatch MCP server delete data?

Yes. 1 tool can delete, cancel or revoke something: remove_recovery_owner.

Can the AgentWatch 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 AgentWatch MCP server have?

21 tools: 1 destructive, 7 write, 13 read. remove_recovery_owner, run_audit, register_intent, add_watch, watch_demo_set, refresh_watches, set_agent_passport, set_recovery_owner and 13 more.

Does the AgentWatch MCP server need sign-in or OAuth?

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

Does the AgentWatch MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the AgentWatch MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 15, 2026.