The AgentWatch MCP server,
connected in one click.
Read-only watchtower for AI agents on-chain: decoded receipts, plan vs execution, Safe audits. Connect the official AgentWatch MCP server to Claude, ChatGPT, or Cursor through gate — one gateway URL, no config files, security-checked.
Free to start · No credit card · No hosting needed
Add AgentWatch to your AI in one minute
Pick your client. Copy, paste, done — or let gate handle sign-ins and policies for you.
claude mcp add --transport http agentwatch https://agentwatch.agentwatch.workers.dev/mcpRun in your terminal, then restart Claude Code.
AgentWatch · https://agentwatch.agentwatch.workers.dev/mcp
What your AI can do with AgentWatch
Real prompts you can type the moment it's connected — no setup, no docs to read.
“Use “list_watched”: List watched addresses for this connector principal. NEW USERS: if empty, call watch_demo_set next (do not invent a global feed), then get_d.”
“Use “get_feed”: Decoded activity feed for an address. Use when reviewing what an agent did on-chain.”
“Use “run_audit”: Run provenance audit on a Safe/address. Use when the user asks 'is this fake?' or wants a shareable verdict.”
“Use “get_capabilities”: Blast-radius / historical capability for an agent key. Use before trusting an agent with funds.”
How secure is the AgentWatch MCP server?
Every server in the gate directory has to pass the same four checks before your AI can touch it. Here's how AgentWatch holds up.
Official endpoint
Operated by AgentWatch (workers.dev) at its documented MCP endpoint — not a third-party mirror or community re-host.
No sign-in needed
This server only exposes public data — no account and no credentials are involved when you connect.
Scanned before it goes live
gate checks every tool for prompt injection, hidden instructions, data-exfiltration hints, and over-broad permissions before the server is available through your gateway.
Watched for drift
If AgentWatch adds or changes tools later, gate re-scans and alerts you — before your AI acts on the change.
Connected in three steps
Connect your gate
Add one gateway URL to Claude, ChatGPT, or any MCP client. One-time setup, about two minutes.
Pick AgentWatch
Choose AgentWatch in the gate directory. No sign-in needed.
Use it anywhere
AgentWatch's tools are live in every AI client you've connected — with rules and logging built in.
Free to start · No credit card · No hosting needed
The actual tools AgentWatch exposes
Fetched live from the official endpoint and re-checked daily by gate — not marketing copy.
list_watched — List watched addresses for this connector principal. NEW USERS: if empty, call watch_demo_set next (do not invent a global feed), then get_d
get_feed — Decoded activity feed for an address. Use when reviewing what an agent did on-chain.
run_audit — Run provenance audit on a Safe/address. Use when the user asks 'is this fake?' or wants a shareable verdict.
get_capabilities — Blast-radius / historical capability for an agent key. Use before trusting an agent with funds.
get_alerts — List alerts scoped to this principal's watched addresses (empty watch list → call watch_demo_set first). Use when checking what needs attent
get_pending_proposals — Pending Safe multisig proposals from the Safe Transaction Service (indexer confidence). Pass safe=0x… or omit to scan all watched addresses.
get_fidelity — Intent↔execution fidelity score for an agent or Safe. Use for accountability / coverage meter. Check matched_agent — false means the address
get_leaderboard — Public standings for agents with a published passport, ranked from receipts (declared-first coverage, fidelity, reliability, hygiene, sustai
+ 13 more tools once connected
Who runs the AgentWatch MCP server?
The AgentWatch MCP server at a glance
https://agentwatch.agentwatch.workers.dev/mcpQuestions, answered
Is there an official AgentWatch MCP server?
Yes — AgentWatch operates its own remote MCP server at agentwatch.agentwatch.workers.dev. gate connects you to that official endpoint and adds security scanning, per-tool access rules, and an activity log on top.
How do I connect AgentWatch to Claude or ChatGPT?
Add gate's gateway URL to your AI client once, then pick AgentWatch in the directory. No account or sign-in is needed. From then on, AgentWatch's tools are available in every client you've connected to gate.
Do I need to host anything?
No. AgentWatch runs the server, and gate is fully managed. You just connect one URL — there's nothing to install, deploy, or maintain.
Is the AgentWatch MCP server safe to connect?
It's the vendor's official endpoint, and gate scans its tools for prompt injection and hidden instructions before it goes live, re-checks it whenever it changes, and lets you allow, block, or require approval for every single tool.
Is it free?
Yes — you can get started with gate for free and connect AgentWatch in minutes. No credit card required.
Connect more than just AgentWatch
Every server below works through the same gateway URL — connect once, add tools anytime.
Squarespace DomainsDesignView →Set up AgentWatch in: Claude · Claude Code · ChatGPT · Cursor · VS Code