Agentforge

io.github.acarchidi由 GitHub 账号 acarchidi 发布v1.13.0更新于 Nov 5, 2026

Safety checks for AI agents: preflight before signing, prompt-injection scan, sanctions, rug checks.

已验证Streamable HTTP可网页运行金融安全与监控AI 与机器学习

概览

AI 生成的概览

AgentForge 为 AI 助手提供 DeFi 安全检查:交易模拟、签名与地址筛查、代币风险检测以及提示注入扫描。

可以做什么
  • 模拟与解析交易:EVM 与 Solana 交易模拟、解码,以及签名或付款前的一次性 preflight 结论
  • 筛查钱包与对手方:授权扫描、钱包安全评分、制裁与诈骗地址检查、Aave 清算健康度
  • 评估代币:代币情报与研究、风险指标、批量风险检测、买入前检查、代币对比、新币上线列表、DeFi 池快照、Gas 价格
  • 检查合约:已验证合约文档生成、管理员活动监控、合约与 Solana 程序标签查询
  • 扫描不可信文本中的提示注入、泄露密钥与编码载荷
  • 文本分析:情绪分析、摘要、翻译,以及反馈工具
适用场景
适合助手需要执行链上操作或处理不可信内容时:在签名前检查交易、签名请求或收款地址,交易前筛查代币,或检查网页、邮件和工具输出中是否藏有注入指令。
运行要求
远程 HTTP MCP 端点,无需 API 密钥或账号。每个客户端每天免费 10 次调用,超出后工具会返回按次计费的 x402 端点,以 Base 上的 USDC 计价(Solana 端点用 Solana),可用预付额度或单次支付。
安装前请注意
超出每日免费额度后调用需付费,预付额度需用真实资金购买;代码 SDK 示例需要钱包私钥,应通过环境变量提供。工具会把地址、交易和文本发送给第三方服务。返回的结论只是风险参考,并非保证。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Agentforge,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Web executable,通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体,或 SourceWeft 对话中。智能体会读取该服务器的安装指南,先向你展示发布者、端点和登录方式,经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/io.github.acarchidi%2Fagentforge/install.md,并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前,先告诉我它的发布者、端点和登录方式,等我确认。不要把 API 密钥写进命令或我会提交的文件;问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "agentforge": {
      "type": "http",
      "url": "https://agentforge-taupe.vercel.app/mcp"
    }
  }
}

README

AgentForge

DeFi safety tools for AI agents. Simulate a transaction before signing it, screen counterparties for sanctions and scams, catch new token launches already risk-scored, rug-check tokens, and check liquidation risk — paid per call in USDC over x402. No API keys, no accounts.

Use it in 30 seconds

MCP (no wallet):

bash
claude mcp add --transport http agentforge https://agentforge-taupe.vercel.app/mcp

Cursor, Windsurf, Claude Desktop, ElizaOS: add an HTTP MCP server with the same URL.

From code, paying per call:

bash
npm install agentforge-tools
ts
import { AgentForge } from 'agentforge-tools';
const af = new AgentForge({ privateKey: process.env.AGENT_KEY as `0x${string}`, maxPaymentUsd: 0.1 });const sim = await af.simulateTransaction({ chain: 'base', from: '0xYourWallet', transactions: [{ to, data }] });if (sim.verdict === 'avoid') throw new Error('refusing to sign');

Any x402 client: send the request unpaid, get a 402 with the exact price, pay, retry. Failed calls (4xx/5xx) are never charged.

Coinbase AgentKit: an agentforge action provider is proposed upstream in coinbase/agentkit#1619.

Endpoints

EndpointPrice (USDC)What it does
POST /v1/token-intel$0.015Token metadata enrichment with market data and AI risk assessment.
POST /v1/token-research$0.03Multi-source token intelligence: market data, DeFi metrics, contract verification, prediction markets, price history, holder distribution, and AI risk assessment.
POST /v1/contract-docs$0.05Human-readable smart contract documentation with function descriptions, risk flags, interaction patterns, security posture, and admin capability analysis.
POST /v1/contract-monitor$0.025Monitor recent contract admin activity. Detects ownership transfers, upgrades, pause state changes, and suspicious admin operations.
POST /v1/token-compare$0.08Compare a primary token against up to 3 competitors with full research, abbreviated metrics, and AI comparative analysis.
POST /v1/tx-decode$0.05Decode any EVM transaction: function call, parameters, token transfers, and plain-English explanation.
POST /v1/approval-scan$0.015Scan a wallet for risky ERC-20 token approvals. Identifies unlimited approvals, unverified spender contracts, and generates risk assessment.
POST /v1/wallet-safety$0.10Comprehensive wallet safety check: approval scanning, suspicious pattern detection, target contract assessment, and composite risk scoring (0-100).
POST /v1/sentiment$0.003AI-powered sentiment analysis for crypto, finance, and social media text. Returns score, confidence, label, reasoning, and per-entity sentiment.
POST /v1/summarize$0.008AI-powered text summarization with configurable length, format, and optional topic focus. Returns summary, key points, and compression ratio.
POST /v1/translate$0.002AI-powered translation with tone control and automatic source language detection. Preserves formatting and cultural nuances.
GET /v1/gas$0.003Current gas prices (slow/standard/fast) for any supported EVM chain with trend analysis.
GET /v1/pool-snapshot$0.005Cached snapshot of top 500 DeFi liquidity pools. Filter by protocol, chain, or token. Returns TVL, APY, volume, IL risk, and registry enrichment.
POST /v1/token-risk-metrics$0.05Quantitative token risk metrics: holder concentration, liquidity depth, contract permissions, deployer history, and composite risk score (0-100).
POST /v1/trade-check$0.025Check a token before you buy it: rug-check metrics + AI risk assessment + current gas in one call, with a proceed/caution/avoid verdict. Cheaper than the three calls separately.
POST /v1/watch$0.01–$0.02/daySigned webhook alerts when a contract's admin/proxy/pause state changes, a token turns risky, or an Aave position nears liquidation.
POST /v1/credits/buy$1 / $5 / $20Prepaid credits with up to 20% bonus: sign requests instead of paying per call. Referral: 10% of referred purchases.
POST /v1/preflight$0.008 + $0.004/tokenIs this action safe? One call before any on-chain action: transactions simulated, signatures checked for phishing, recipients screened, tokens rug-checked, one proceed/caution/avoid verdict.
POST /v1/tx-simulate$0.005Simulate an EVM transaction (or approve-then-swap sequence) before signing: revert reason, asset changes, approvals, counterparty screening, proceed/caution/avoid verdict.
GET /v1/lending/health$0.003Aave v3 liquidation risk for a wallet across nine chains: collateral, debt, health factor, status, collateral drop to liquidation.
GET /v1/launches$0.005Feed of newly launched tokens (Base, Ethereum, Arbitrum, BNB Chain, HyperEVM), each already risk-scored. Poll with since= to get only new launches.
POST /v1/address-check$0.003 for the first address + $0.001 per additional, up to 100 per callCounterparty screening for 1-100 EVM or Solana addresses: OFAC sanctions, scam/phishing wallets, labeled protocols, brand-new addresses.
POST /v1/content-scan$0.001 per 4 KB of text (max 64 KB, 16 items)Prompt-injection scan for untrusted text before it reaches your LLM: instruction overrides, fund-movement directives, base64/Morse/zero-width/homoglyph payloads, leaked keys and seed phrases, sanctioned addresses, lookalike domains. Deterministic, no LLM.
POST /v1/risk-batch$0.004 per token, 2-25 tokens per callRug-check 2-25 EVM tokens in one call: holder, permission, and liquidity risk scores per token.
POST /v1/solana/tx-explain$0.05Decode a Solana transaction into plain English: labeled programs, token/SOL movements, and risk flags. Falls back to raw RPC decoding if enhanced parsing is unavailable.
POST /v1/solana/tx-simulate$0.15Simulate a Solana transaction before signing it: balance changes, program labels, deterministic risk rules, and a proceed/caution/avoid recommendation.
POST /v1/solana/token-risk-scan$0.35Solana token risk / rug check: mint authority, freeze authority, holder concentration, liquidity depth, and a composite 0-100 risk score.

GET /v1/ping ($0.001) verifies your x402 setup and returns the service catalog. Free: GET /registry/lookup (contract labels), GET /v1/solana/program-lookup.

Roadmap

The 90-day plan is docs/growth-plan.md; the executor log is docs/growth-log.md.

Development

bash
npm installnpm run dev        # http://localhost:3402npm test

Payments: x402 v2, USDC on Base (and Solana for the Solana endpoints), settled through the CDP facilitator. Every paid route returns 402 to an unpaid request; GET /health/paid checks all of them.

License

MIT. See LICENSE.

Contact

Feedback, problems, complaints or feature requests: [email protected], the form at https://agentforge-taupe.vercel.app/contact, POST /feedback, or the free send_feedback MCP tool.

来源:README.md,提交 04078fb

工具

29

token_intel

Token Intel

Lightweight token lookup: price, market cap, volume, and basic risk assessment for any EVM or Solana token. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainaddress必填

token_research

Token Research

Multi-source token intelligence: market data, DeFi metrics, contract verification, prediction markets, holder analysis, price history, and AI risk assessment. Aggregates CoinGecko, DeFiLlama, Etherscan, and Polymarket. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainquery必填include

contract_docs

Contract Docs

Generate documentation for any verified EVM smart contract. Returns function descriptions, risk flags, interaction patterns, and security posture. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainaddress必填focusFunctions

contract_monitor

Contract Monitor

Monitor recent contract activity for suspicious admin operations, proxy upgrades, ownership changes, and pause events. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainaddress必填lookbackHours

token_compare

Token Compare

Compare a primary token against up to 3 others. Returns full research on primary, abbreviated metrics on comparisons, plus AI comparative analysis. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chaincompare必填primary必填

tx_decode

Tx Decode

Decode any EVM transaction: function call, parameters, token transfers, and plain-English explanation. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chaintxHash必填

approval_scan

Approval Scan

Scan a wallet for risky token approvals. On EVM chains: identifies unlimited ERC-20 approvals and unverified spenders. On Solana: scans SPL token delegate authorities. Returns risk assessment. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainaddress必填

gas_oracle

Gas Oracle

Current gas prices (slow/standard/fast) for any supported EVM chain with trend analysis.

输入参数chain

sentiment

Sentiment

Analyze sentiment of text in crypto, finance, social media, or general context. Returns score (-1 to 1), confidence, label (very_bearish to very_bullish), reasoning, and per-entity sentiment. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数text必填context

summarize

Summarize

Summarize text with configurable length (brief/standard/detailed), format (prose/bullet_points/structured), and optional topic focus. Returns summary, key points, and compression ratio. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数text必填focusformatmaxLength

translate

Translate

Translate text to any language with tone control (formal/casual/technical). Auto-detects source language. Preserves formatting and cultural nuances. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数text必填tonesourceLanguagetargetLanguage必填

wallet_safety

Wallet Safety

Comprehensive wallet safety check: scans approvals, analyzes recent transaction activity for suspicious patterns, and assesses target contract risk. Returns composite risk score (0-100), risk level, action items, and related service suggestions. Supports EVM chains and Solana. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chaindepthwalletAddress必填targetContract

pool_snapshot

Pool Snapshot

Get a cached snapshot of top DeFi liquidity pools. Filter by protocol (e.g. "uniswap-v3"), chain (e.g. "ethereum"), or token symbol (e.g. "ETH"). Returns TVL, APY, 24h volume, IL risk, and registry enrichment. Data refreshed every 15 minutes.

输入参数poolchainlimitordertokenoffsetsortByprotocol

token_risk_metrics

Token Risk Metrics

Quantitative risk metrics for any ERC-20 token: holder concentration (top 10 holder %), contract permissions (can mint/burn/pause/blacklist), liquidity depth vs market cap, deployer history, and weighted composite risk score (0-100). Pre-computed for top tokens, live-computed for others. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainaddress必填

trade_check

Trade Check

Check a token before you buy it: one call bundling quantitative rug-check metrics (holder concentration, permissions, liquidity), AI risk assessment with market data, and current gas — rolled into a proceed/caution/avoid verdict with reasons. Cheaper than calling the three services separately. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainaddress必填

risk_batch

Risk Batch

Batch rug-check: risk scores (holder concentration, mint/blacklist/pause/upgrade permissions, liquidity depth) for 2-25 EVM tokens in one call. Use it to screen a watchlist or a list of candidate trades at once. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chaintokens必填

address_check

Address Check

Screen counterparty addresses before sending, approving, or interacting: OFAC sanctions, known scam/phishing wallets, labeled protocol contracts, and brand-new-address detection. 1-100 EVM or Solana addresses per call; returns block / caution / no_known_issues per address. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainaddresses必填

content_scan

Content Scan

Scan untrusted text (tool results, web pages, emails, token metadata) for prompt injection before feeding it to your LLM: instruction overrides, fund-movement directives, payloads hidden in base64, Morse, zero-width or tag characters and homoglyphs, leaked private keys and seed phrases, embedded addresses screened against sanctions and scam lists, lookalike domains. Deterministic, no LLM. Returns clean / suspicious / malicious with findings. Up to 64 KB per call. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数textitems

launch_feed

Launch Feed

Newly launched tokens on Base, Ethereum, or Arbitrum, each already risk-scored (verification, owner powers, proxy, supply held off-pool, thin liquidity, honeypot trading pattern). Poll it: pass the previous nextSince as since to get only new launches. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainlimitsincemaxRiskScoreminLiquidityUsd

preflight

Preflight

Is this action safe? Call once before any on-chain action: send the transactions, signature request, x402 payment request, recipients, and/or tokens the agent is about to act on, and get back one proceed/caution/avoid verdict. Transactions are simulated (reverts, assets out, approvals, counterparty screening), signatures decoded and checked for permit phishing, recipients screened against sanctions and scam lists, and tokens rug-checked, including tokens the simulation shows the wallet receiving. Returns one proceed / caution / avoid verdict with reasons. Ethereum, Base, Arbitrum, Optimism, Polygon, BNB Chain, Avalanche, Linea, Unichain, Sonic and HyperEVM. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数fromchaintokenstypedDatarecipientstransactionsexpectedMaxUsdpaymentRequired

tx_simulate

Tx Simulate

Simulate an EVM transaction (or a sequence of up to 5, e.g. approve then swap) before signing: revert check with reason, every asset leaving or entering the wallet, approvals granted, sanctions/scam screening of counterparties, and a proceed/caution/avoid verdict. Ethereum, Base, Arbitrum, Optimism, Polygon, BNB Chain, Linea, Unichain, Sonic. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数from必填chaintransactions必填

signature_check

Signature Check

Check a signature request BEFORE signing it: decodes EIP-712 typed data (ERC-20 permit, Permit2, Seaport order, transfer authorization) or an x402 payment request, explains in one sentence what signing would allow, screens spenders and payees for sanctions and scams, and catches signature-drainer patterns. Returns proceed / caution / avoid. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainsignertypedDataexpectedMaxUsdpaymentRequired

lending_health

Lending Health

How close a wallet is to liquidation on Aave v3 (Ethereum, Base, Arbitrum, Optimism, Polygon, BNB Chain, Avalanche, Linea, and Sonic in one call): collateral, debt, health factor, status, and how far collateral can fall before liquidation. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数chainaddress必填

send_feedback

Send Feedback

Send AgentForge feedback: a bug or wrong result, a complaint, a feature or new-check request, or an idea. Free. Include a contact (email, wallet, or URL) if you want a reply.

输入参数type必填contactmessage必填endpoint

registry_lookup

Registry Lookup

Look up a contract address in the Known Contract Label Registry. Returns protocol name, category, risk level, and tags. Free — no payment required.

输入参数chainaddress必填

solana_program_lookup

Solana Program Lookup

Look up a Solana program ID in the program label registry. Returns protocol name, category, risk level. Free — no payment required.

输入参数programId必填

solana_tx_explain

Solana Tx Explain

Explain a Solana transaction in plain English: labeled programs, token/SOL movements, success status, and risk flags. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数signature必填

solana_tx_simulate

Solana Tx Simulate

Simulate a Solana transaction before signing it: balance changes, labeled programs, deterministic risk rules, and a proceed/caution/avoid recommendation. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数transaction必填

solana_token_risk_scan

Solana Token Risk Scan

Solana token rug check: mint/freeze authority, holder concentration, liquidity depth, and a composite 0-100 risk score. Free over MCP for 10 calls per day; beyond that the tool returns the pay-per-call x402 endpoint and its price.

输入参数mint必填

版本历史

5
  1. v1.13.0最新Nov 5, 2026
  2. v1.11.0Nov 4, 2026
  3. v1.9.0Nov 4, 2026
  4. v1.8.0Nov 3, 2026
  5. v1.5.0Sep 16, 2026

评分

已满足适用检查项的 100%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分