AgentWatch

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

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

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "AgentWatch": {
      "url": "https://agentwatch.agentwatch.workers.dev/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (21)

• 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_dashboard.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_feed

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

  Parameters:

  • address (required) — string — Watched address (0x…40 hex)
  • since — number — Unix seconds lower bound

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• run_audit

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

  Parameters:

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

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_capabilities

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

  Parameters:

  • agent_key (required) — string

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_alerts

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

  Parameters:

  • status — string

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_pending_proposals

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

  Parameters:

  • safe — string — Optional Safe address; default = all watched
  • chains — string — Optional comma-separated chain ids

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_fidelity

  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.

  Parameters:

  • agent_key (required) — string — Agent EOA address (0x…40 hex)

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_leaderboard

  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.

  Parameters:

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

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_benchmark

  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.

  Parameters:

  • agent_key (required) — string — Agent EOA address (0x…40 hex)

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_tca

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

  Parameters:

  • event_id (required) — string

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• register_intent

  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.

  Parameters:

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

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• add_watch

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

  Parameters:

  • address (required) — string — Address to watch (0x…40 hex)
  • label — string — Optional short label (e.g. Agent EOA)
  • chains — string

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• watch_demo_set

  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.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• refresh_watches

  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.

  Parameters:

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

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_agent_passport

  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.

  Parameters:

  • address (required) — string — Watched agent address (0x…40 hex)

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• set_agent_passport

  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.

  Parameters:

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

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• list_recovery_owners

  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.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• set_recovery_owner

  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.

  Parameters:

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

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• remove_recovery_owner

  Remove a saved recovery address for this user.

  Parameters:

  • address (required) — string

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_safe_defaults

  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.

  Parameters:

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

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_dashboard

  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.

  Parameters:

  • persona — string — Optional override; otherwise uses saved persona

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

Publisher website

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.