// SCAN REPORT

Agent

9 tools. 3 can modify or destroy data without limits.

2 destructive tools with no built-in limits. Policy required.

Last updated: 10 Jul 2026

3 can modify or destroy data

6 read-only

9 tools total

Community server · catalogue entry checked 10/07/2026 · full schemas captured for 8 of 9 tools

How to control Agent ↓

TOOL MAP

What Agent exposes to your agents

Read (6) Write / Execute (1) Destructive / Financial (2)

Running more than one server? Check your whole stack →

CONTEXT-WINDOW COST

What Agent costs in tokens

1,663 tokens of tool definitions, loaded on every request

0.8% of a 200k context window

247 heaviest tool: build_swap

Full per-tool breakdown → Model it in the token calculator →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Agent tools

Financial · High build_lp_deposit Although the tool only builds the transaction and does not broadcast it, the resulting artifact is a financial operation that moves user tokens into a liquidity pool. Financial · High build_lp_withdraw While the tool itself only builds the transaction (no broadcast), it constructs a complete financial operation to withdraw liquidity pool assets. Execute · Medium build_swap Build a NON-CUSTODIAL ERC20 swap plan (approve → Permit2 sign → router.swap) for the caller to sign + broadcast. No broadcast.

3 of Agent's 9 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Agent

PolicyLayer is an MCP gateway — it sits between your AI agents and Agent, and nothing reaches the server without passing your rules. These are the rules we recommend:

Block financial tools by default

{
  "build_lp_deposit": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Requires human approval."
      }
    ]
  }
}

Financial tools should be explicitly enabled per use case, not open by default.

Cap read operations

{
  "list_tokens": {
    "limits": [
      {
        "counter": "list_tokens_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Agent — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON AGENT →

Instant setup, no code required.

FULL CATALOGUE

All 9 Agent tools

FINANCIAL 2 tools

Financial build_lp_deposit Build a NON-CUSTODIAL any-token LP deposit plan (approve → Permit2 sign → router.depositAnyToken) for the call Financial build_lp_withdraw Build a NON-CUSTODIAL any-token LP withdraw plan (approve shares → Permit2 sign → router.withdrawAnyToken) for

EXECUTE 1 tools

Execute build_swap Build a NON-CUSTODIAL ERC20 swap plan (approve → Permit2 sign → router.swap) for the caller to sign + broadcas

READ 6 tools

Read list_tokens List the public swap token allow-list (address/symbol/name/decimals). Public market data only — no user-privat Read list_vaults List the public any-token LP vaults (addresses + token pair). Public market data only — no user-private reads. Read ping Health check — returns 'pong'. Read quote_lp_deposit Read-only any-token LP deposit preview: estimated shares, guaranteed minShares, fee, and refund bps. No broadc Read quote_lp_withdraw Read-only any-token LP withdraw preview: estimated and guaranteed token-out for burning vault shares. No broad Read quote_swap Read-only ERC20 swap quote: expected out, guaranteed minOut, fee, and a high-level route summary. No broadcast

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Agent

Can an AI agent move money through the Agent MCP server? +

Yes. The Agent server exposes 2 financial tools including build_lp_deposit, build_lp_withdraw. Without a policy, an autonomous agent can call these with no spend caps, no rate limits, and no approval flow. PolicyLayer lets you block financial tools by default, require human approval, or set per-tool rate limits — enforced on every call.

How many tools does the Agent MCP server expose? +

9 tools across 4 categories: Execute, Financial, Other, Read. 6 are read-only. 3 can modify, create, or delete data.

How do I enforce a policy on Agent? +

Register the Agent MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Agent tool call.

Deterministic rules across all 9 Agent tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON AGENT →

Instant setup, no code required.

9 Agent tools catalogued and risk-classified — across an index of 46,500+ MCP servers.

// WHERE THIS COMES FROM

These policies come from Agent's registry record.

The record behind this page: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.

Teams ship this data inside their own products. See what a licence covers →