@sperax/tool-token-approvals

Scan, assess, and revoke token approvals to protect your wallet from unlimited allowances

Token Approvals is an agent tool from SperaxOS, packaged headless so you can call it from any agent framework. It ships two things: the manifest — a JSON-Schema function definition a model can call — and the executor that runs the call against the real API.

There is no UI layer and no framework lock-in. It works anywhere TypeScript runs.

Install

npm install @sperax/tool-token-approvals

Usage

Call it directly

import { tokenApprovalsExecutor } from '@sperax/tool-token-approvals';

const result = await tokenApprovalsExecutor.invoke('getApprovals', {"address":"<address>","chain":"<chain>"}, {
  messageId: 'msg-1',
});

console.log(result.content); // prose summary written for the model to read
console.log(result.state);   // typed data payload for your own UI

Give it to a model

import Anthropic from '@anthropic-ai/sdk';
import { TokenApprovalsManifest, tokenApprovalsExecutor } from '@sperax/tool-token-approvals';

const client = new Anthropic();

const response = await client.messages.create({
  model: 'claude-opus-4-8',
  max_tokens: 1024,
  messages: [{ role: 'user', content: 'Ask something this tool can answer' }],
  tools: TokenApprovalsManifest.api.map((api) => ({
    name: api.name,
    description: api.description,
    input_schema: api.parameters,
  })),
});

for (const block of response.content) {
  if (block.type !== 'tool_use') continue;
  const result = await tokenApprovalsExecutor.invoke(block.name, block.input, { messageId: response.id });
  console.log(result.content);
}

TokenApprovalsManifest.api is already in JSON-Schema form, so it maps onto any tool-calling API — Anthropic, OpenAI, the Vercel AI SDK, or an MCP server — without translation.

Every executor returns a BuiltinToolResult — { success, content, state }. content is prose written for the model to read; state is the typed data payload for your own code. Executors never throw: a failed call comes back as { success: false, content: '<reason>' }, so a network blip degrades the answer instead of crashing the agent loop.

Configuration — required

This tool needs a backend you control. It will not work on a bare npm install alone.

The upstream API requires a secret key. That key is deliberately not bundled here — shipping it in an npm package would leak it to every consumer. Instead the executor calls a SperaxOS /webapi/* route, which holds the key server-side and injects it. That route is session-authenticated, so the public deployment at https://chat.sperax.io (the default origin) answers 401 to anonymous callers.

To use this tool you need one of:

• a SperaxOS deployment of your own, or
• any HTTP endpoint that implements the same request shape and supplies the key.

Point the package at it before importing the tool — the URL is resolved once, when the module first loads:

SPERAX_API_BASE_URL=https://my-speraxos.example.com

or in code:

import { configureSperaxApi } from '@sperax/agent-tools-core';

configureSperaxApi({ baseUrl: 'https://my-speraxos.example.com' });

// import the tool only after configuring, so the path resolves against your origin
const { tokenApprovalsExecutor } = await import('@sperax/tool-token-approvals');

Inside a browser that already serves those routes at its own origin, requests stay same-origin and no configuration is needed.

If you want a tool that runs with zero setup, use one of the standalone tools — those call public APIs directly and need no key, no origin, and no backend.

Tool identifier

sperax-token-approvals

API reference

getApprovals

Scan all active ERC-20 and ERC-721 token approvals (allowances) for a wallet address. Shows which contracts can spend your tokens, the approved amounts, and risk levels. Displays a rich approval list with risk indicators.

| Parameter | Type | Required | Description | | --- | --- | --- | --- | | address | string | yes | Wallet address (0x…) | | chain | string | no | Chain name (e.g. "arbitrum", "ethereum", "polygon"). Defaults to arbitrum. | | tokenType | string | no | Filter by token type: "ERC-20", "ERC-721", or "all" (default: "all") |

getRiskAssessment

Get a comprehensive security risk assessment of all token approvals for a wallet. Returns an overall risk score, counts of unlimited/unverified approvals, total value at risk, and specific risk findings.

| Parameter | Type | Required | Description | | --- | --- | --- | --- | | address | string | yes | Wallet address (0x…) | | chain | string | no | Chain name (e.g. "arbitrum", "ethereum", "polygon"). Defaults to arbitrum. |

revokeApproval

Build an unsigned revoke transaction for a specific token approval. Returns transaction data (to, data, chainId) that the user must sign with their wallet. For ERC-20: calls approve(spender, 0). For ERC-721: calls setApprovalForAll(spender, false).

| Parameter | Type | Required | Description | | --- | --- | --- | --- | | address | string | yes | Wallet address (0x…) | | chain | string | no | Chain name (e.g. "arbitrum", "ethereum", "polygon"). Defaults to arbitrum. | | spenderAddress | string | yes | Spender contract address to revoke approval from | | tokenAddress | string | yes | Token contract address | | tokenType | string | no | Token type: "ERC-20" or "ERC-721" (default: "ERC-20") |

revokeAllApprovals

Build unsigned revoke transactions for all active approvals (or only high-risk ones). Returns an array of transactions that the user can sign in batch. Use riskyOnly=true to only revoke unlimited and high-risk approvals.

| Parameter | Type | Required | Description | | --- | --- | --- | --- | | address | string | yes | Wallet address (0x…) | | chain | string | no | Chain name (e.g. "arbitrum", "ethereum", "polygon"). Defaults to arbitrum. | | riskyOnly | boolean | no | If true, only revoke high/critical risk approvals. If false, revoke all. Default: true. |

getApprovalHistory

View the history of approval and revoke events for a wallet. Shows when approvals were granted and revoked, with timestamps and transaction hashes.

| Parameter | Type | Required | Description | | --- | --- | --- | --- | | address | string | yes | Wallet address (0x…) | | chain | string | no | Chain name (e.g. "arbitrum", "ethereum", "polygon"). Defaults to arbitrum. | | limit | number | no | Maximum number of events to return (default: 20) |

Types

Shared types come from @sperax/agent-tools-core: BuiltinToolManifest, BuiltinToolResult, BuiltinToolContext, and the BaseExecutor class every tool executor extends.

Related

• @sperax/agent-tools-core — the tool contract
• All SperaxOS agent tools — tool-token-approvals is one of many
• SperaxOS — the agent workspace these tools were built for

License

Apache-2.0