# AgentWatch (remote · agentwatch.agentwatch.workers.dev)

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

- Trust score: 83/100 (high trust)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `agentwatch.agentwatch.workers.dev`: 83/100 (this document), [markdown](https://verifymcp.io/servers/dev-workers-agentwatch-agentwatch-agentwatch/agentwatch.md), [page](https://verifymcp.io/servers/dev-workers-agentwatch-agentwatch-agentwatch/agentwatch)

## Channel facts

- Endpoint: `https://agentwatch.agentwatch.workers.dev/mcp`
- Transports: `streamable-http`
- Auth: `required`
- Version: `0.5.1`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Endpoint Security**: 78/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
  - HTTPS check failed: the endpoint is reachable over plaintext HTTP.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
  - The authorisation server offers only Dynamic Client Registration (RFC 7591), which MCP 2026-07-28 deprecated in favour of Client ID Metadata Documents.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 86/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (excellent).
  - Tool/resource definitions use about 1907 tokens (~86/item across 22 items; 21 tools + 1 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 97/100
  - Stability observed for 29 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 93/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 78% of tool parameters carry a description.
- **Tool Safety**: 50/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "remove_recovery_owner" implies "remove" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - Manipulation not yet verified: none of the 23 captured unit(s) of tool text has been judged yet, so we will not certify text no model has read as clean.
- **Capabilities**: 60/100
  - Spec-recency check failed: implements MCP spec 2025-03-26; the latest is 2026-07-28.
  - Supports UI / widget rendering.

## Install

### How do I install the AgentWatch MCP server?

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

### Claude

```bash
claude mcp add --transport http dev-workers-agentwatch-agentwatch-agentwatch 'https://agentwatch.agentwatch.workers.dev/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "dev-workers-agentwatch-agentwatch-agentwatch": {
      "url": "https://agentwatch.agentwatch.workers.dev/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "dev-workers-agentwatch-agentwatch-agentwatch": {
      "type": "http",
      "url": "https://agentwatch.agentwatch.workers.dev/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.dev-workers-agentwatch-agentwatch-agentwatch]
url = "https://agentwatch.agentwatch.workers.dev/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "dev-workers-agentwatch-agentwatch-agentwatch": {
      "type": "remote",
      "url": "https://agentwatch.agentwatch.workers.dev/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add dev-workers-agentwatch-agentwatch-agentwatch --url 'https://agentwatch.agentwatch.workers.dev/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  dev-workers-agentwatch-agentwatch-agentwatch:
    url: "https://agentwatch.agentwatch.workers.dev/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "dev-workers-agentwatch-agentwatch-agentwatch": {
      "Transport": "http",
      "Url": "https://agentwatch.agentwatch.workers.dev/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add dev-workers-agentwatch-agentwatch-agentwatch -t streamable-http -u 'https://agentwatch.agentwatch.workers.dev/mcp'
```

### Other

```json
{
  "mcpServers": {
    "dev-workers-agentwatch-agentwatch-agentwatch": {
      "type": "http",
      "url": "https://agentwatch.agentwatch.workers.dev/mcp"
    }
  }
}
```

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

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-04 (score 83, +1)

No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-02 (score 82, +1)

No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-31 (score 81, +1)

No change was recorded against any check on this day. Stability & Change Management went from 77 to 80. That category is still filling its 30-day observation window: 23 days of observed history at the previous scan, 24 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-29 (score 80, +1)

No change was recorded against any check on this day. Stability & Change Management went from 70 to 73. That category is still filling its 30-day observation window: 21 days of observed history at the previous scan, 22 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-27 (score 79, +1)

No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-24 (score 78, +1)

No change was recorded against any check on this day. Stability & Change Management went from 53 to 57. That category is still filling its 30-day observation window: 16 days of observed history at the previous scan, 17 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-22 (score 77, +1)

No change was recorded against any check on this day. Stability & Change Management went from 47 to 50. That category is still filling its 30-day observation window: 14 days of observed history at the previous scan, 15 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-20 (score 76, +1)

No change was recorded against any check on this day. Stability & Change Management went from 40 to 43. That category is still filling its 30-day observation window: 12 days of observed history at the previous scan, 13 at this one. The score rises as the window fills, whether or not the server changes.

## MCP tools (21)

### `list_watched` (~40 tokens)

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.

### `get_feed` (~53 tokens)

Decoded activity feed for an address. Use when reviewing what an agent did on-chain.

Input parameters:

- `address` (string, required): Watched address (0x…40 hex)
- `since` (number): Unix seconds lower bound

### `run_audit` (~70 tokens)

Run provenance audit on a Safe/address. Use when the user asks 'is this fake?' or wants a shareable verdict.

Input parameters:

- `address` (string, required): Safe/address to audit (0x…40 hex)
- `agents` (string): Comma-separated agent keys
- `chain` (string)

### `get_capabilities` (~35 tokens)

Blast-radius / historical capability for an agent key. Use before trusting an agent with funds.

Input parameters:

- `agent_key` (string, required)

### `get_alerts` (~43 tokens)

List alerts scoped to this principal's watched addresses (empty watch list → call watch_demo_set first). Use when checking what needs attention.

Input parameters:

- `status` (string)

### `get_pending_proposals` (~66 tokens)

Pending Safe multisig proposals from the Safe Transaction Service (indexer confidence). Pass safe=0x… or omit to scan all watched addresses.

Input parameters:

- `chains` (string): Optional comma-separated chain ids
- `safe` (string): Optional Safe address; default = all watched

### `get_fidelity` (~75 tokens)

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.

Input parameters:

- `agent_key` (string, required): Agent EOA address (0x…40 hex)

### `get_leaderboard` (~108 tokens)

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.

Input parameters:

- `agent_key` (string): Optional: address to locate in the standings
- `limit` (number): How many ranked rows to return (default 10, max 50)

### `get_benchmark` (~62 tokens)

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.

Input parameters:

- `agent_key` (string, required): Agent EOA address (0x…40 hex)

### `get_tca` (~38 tokens)

TCA market-context snapshot for an event_id (chain:txHash). Use when judging execution quality.

Input parameters:

- `event_id` (string, required)

### `register_intent` (~199 tokens)

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.

Input parameters:

- `agent_key` (string, required)
- `deadline_minutes` (number): Relative window; stored as absolute constraints.deadline
- `declared_at` (number)
- `description` (string, required)
- `expected` (object): Structured shape. Use steps: ['approve','other'] (or similar) so one intent covers approve+supply/swap companions.

### `add_watch` (~62 tokens)

Start watching an address. Use when onboarding a new agent or Safe.

Input parameters:

- `address` (string, required): Address to watch (0x…40 hex)
- `chains` (string)
- `label` (string): Optional short label (e.g. Agent EOA)

### `watch_demo_set` (~96 tokens)

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.

### `refresh_watches` (~96 tokens)

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.

Input parameters:

- `address` (string): Optional single address to refresh (0x…40 hex)
- `chains` (string): Optional csv of chains (default: ethereum,base first, then the rest)

### `get_agent_passport` (~61 tokens)

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.

Input parameters:

- `address` (string, required): Watched agent address (0x…40 hex)

### `set_agent_passport` (~190 tokens)

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.

Input parameters:

- `address` (string, required): Watched agent address (0x…40 hex)
- `cadence` (string): always_on, scheduled, event_driven, on_demand
- `display_name` (string): Name shown instead of the raw address
- `model` (string): Optional model or framework detail
- `operator` (string): Who runs it
- `provider` (string): Which model drives the agent
- `public` (boolean): Publish a public passport page
- `purpose` (string): One or two sentences on what it is for
- `role` (string)

### `list_recovery_owners` (~48 tokens)

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.

### `set_recovery_owner` (~122 tokens)

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.

Input parameters:

- `address` (string, required): 0x recovery wallet the human controls
- `label` (string): Short name, e.g. hardware, treasury, personal
- `make_default` (boolean): Prefer this label for default Safes
- `purpose` (string): What this recovery is for, e.g. default, trading

### `remove_recovery_owner` (~26 tokens)

Remove a saved recovery address for this user.

Input parameters:

- `address` (string, required)

### `get_safe_defaults` (~148 tokens)

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.

Input parameters:

- `agent_key` (string, required): Agent EOA that will be an owner/signer
- `all_recoveries` (boolean): If true, include all matching recoveries (1/n) instead of just the default one
- `chain` (string): Optional chain hint for the intent constraints
- `purpose` (string): Optional purpose filter (e.g. trading)
- `recovery_label` (string): Optional label to pick among several recoveries

### `get_dashboard` (~84 tokens)

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.

Input parameters:

- `persona` (string): Optional override; otherwise uses saved persona

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/dev-workers-agentwatch-agentwatch-agentwatch/agentwatch#diagnostics

## Score history

- 2026-11-05: 83
- 2026-11-04: 83
- 2026-11-03: 82
- 2026-11-02: 82
- 2026-11-01: 81
- 2026-10-31: 81
- 2026-10-30: 80
- 2026-10-29: 80
- 2026-10-28: 79
- 2026-10-27: 79
- 2026-10-26: 78
- 2026-10-25: 78
- 2026-10-24: 78
- 2026-10-23: 77
- 2026-10-22: 77
- 2026-10-21: 76
- 2026-10-20: 76
- 2026-10-19: 75
- 2026-10-18: 75
- 2026-10-17: 74
- 2026-10-16: 74
- 2026-10-15: 71
- 2026-10-14: 73
- 2026-10-13: 72
- 2026-10-12: 72
- 2026-10-11: 71
- 2026-10-10: 71
- 2026-10-08: 70

## Common questions

### What is the AgentWatch MCP server?

AgentWatch is an MCP server listed in the public MCP registry as dev.workers.agentwatch.agentwatch/agentwatch. Read-only watchtower for AI agents on-chain: decoded receipts, plan vs execution, Safe audits. This page covers its hosted endpoint (https://agentwatch.agentwatch.workers.dev/mcp).

### Is the AgentWatch MCP server safe to use?

AgentWatch scores 83 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the AgentWatch MCP server expose?

AgentWatch exposes 21 tools: list_watched, get_feed, run_audit, get_capabilities, get_alerts, and 16 more. Their descriptions and schemas cost roughly 1,722 tokens of context every time the server is loaded.

### Does the AgentWatch MCP server require authentication?

Yes. AgentWatch asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.

### Is the AgentWatch MCP server still maintained?

AgentWatch is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://agentwatch.agentwatch.workers.dev/mcp
- Website: https://agentwatch.agentwatch.workers.dev/
- Changelog RSS feed: https://verifymcp.io/servers/dev-workers-agentwatch-agentwatch-agentwatch/agentwatch.xml
- Changelog JSON feed: https://verifymcp.io/servers/dev-workers-agentwatch-agentwatch-agentwatch/agentwatch.json
- HTML version of this page: https://verifymcp.io/servers/dev-workers-agentwatch-agentwatch-agentwatch/agentwatch