VulnHunt Security Intelligence
Read-only smart-contract security intelligence for autonomous agents.
我该使用它吗
质量与安全性
B
发现（5）
- LOWTool 'vulnhunt_decision' description lacks action verb在 vulnhunt_decision 中
- LOWTool 'gpufeed_compare_providers' description lacks action verb在 gpufeed_compare_providers 中
- LOWTool 'gpufeed_status' description lacks action verb在 gpufeed_status 中
- LOWTool 'gpufeed_get_history' description lacks action verb在 gpufeed_get_history 中
- LOWTool 'gpufeed_get_price_index' description lacks action verb在 gpufeed_get_price_index 中
基于对工具定义和协议合规性的自动分析。
上下文开销
~1,047token 数（工具定义）
~803 B典型响应大小
对注意力有中等影响（占 128k 上下文窗口的 0.82%）
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高，可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中：
{
"mcpServers": {
"vulnhunt": {
"url": "https://vulnhunt-agent-api-production.up.railway.app/mcp"
}
}
}远程端点
https://vulnhunt-agent-api-production.up.railway.app/mcpstreamable-http它能做什么
工具清单
工具（11）
🟢 只读🟡 写入🔴 删除⚪ 未知
⚪vulnhunt_decision(chain_id, address)
Return a bounded free decision packet for an EVM contract.
输入模式
{
"type": "object",
"properties": {
"chain_id": {
"title": "Chain Id",
"type": "integer"
},
"address": {
"title": "Address",
"type": "string"
}
},
"required": [
"chain_id",
"address"
],
"title": "vulnhunt_decisionArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "vulnhunt_decisionDictOutput"
}⚪vulnhunt_payment_resources
Return the machine-readable paid resource catalog and x402 terms.
输入模式
{
"type": "object",
"properties": {},
"title": "vulnhunt_payment_resourcesArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "vulnhunt_payment_resourcesDictOutput"
}🟢vulnhunt_chains
List supported chains.
输入模式
{
"type": "object",
"properties": {},
"title": "vulnhunt_chainsArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "vulnhunt_chainsDictOutput"
}🟢gpufeed_get_cheapest(gpu, count, hours)
Return the cheapest verified free-first GPU offer.
输入模式
{
"type": "object",
"properties": {
"gpu": {
"title": "Gpu",
"type": "string"
},
"count": {
"default": 1,
"title": "Count",
"type": "integer"
},
"hours": {
"default": 1,
"title": "Hours",
"type": "number"
}
},
"required": [
"gpu"
],
"title": "gpufeed_get_cheapestArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "gpufeed_get_cheapestDictOutput"
}🟢gpufeed_get_availability(gpu, count)
Return observed GPU listings from free-first sources.
输入模式
{
"type": "object",
"properties": {
"gpu": {
"title": "Gpu",
"type": "string"
},
"count": {
"default": 1,
"title": "Count",
"type": "integer"
}
},
"required": [
"gpu"
],
"title": "gpufeed_get_availabilityArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "gpufeed_get_availabilityDictOutput"
}⚪gpufeed_compare_providers(gpu, providers)
Compare providers without treating missing data as unavailable.
输入模式
{
"type": "object",
"properties": {
"gpu": {
"title": "Gpu",
"type": "string"
},
"providers": {
"default": "",
"title": "Providers",
"type": "string"
}
},
"required": [
"gpu"
],
"title": "gpufeed_compare_providersArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "gpufeed_compare_providersDictOutput"
}⚪gpufeed_status
Return GPUFEED scope, guardrail, metering and provider status.
输入模式
{
"type": "object",
"properties": {},
"title": "gpufeed_statusArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "gpufeed_statusDictOutput"
}🟢gpufeed_get_history(gpu, days, provider, limit)
Return persisted verified GPU price observations.
输入模式
{
"type": "object",
"properties": {
"gpu": {
"title": "Gpu",
"type": "string"
},
"days": {
"default": 7,
"title": "Days",
"type": "integer"
},
"provider": {
"default": "",
"title": "Provider",
"type": "string"
},
"limit": {
"default": 100,
"title": "Limit",
"type": "integer"
}
},
"required": [
"gpu"
],
"title": "gpufeed_get_historyArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "gpufeed_get_historyDictOutput"
}🟢gpufeed_get_price_index(gpu, days)
Return descriptive statistics over persisted GPU observations.
输入模式
{
"type": "object",
"properties": {
"gpu": {
"title": "Gpu",
"type": "string"
},
"days": {
"default": 7,
"title": "Days",
"type": "integer"
}
},
"required": [
"gpu"
],
"title": "gpufeed_get_price_indexArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "gpufeed_get_price_indexDictOutput"
}🟢gpufeed_estimate_training(gpu, hours, count)
Estimate GPU compute cost using observed prices, with assumptions exposed.
输入模式
{
"type": "object",
"properties": {
"gpu": {
"title": "Gpu",
"type": "string"
},
"hours": {
"title": "Hours",
"type": "number"
},
"count": {
"default": 1,
"title": "Count",
"type": "integer"
}
},
"required": [
"gpu",
"hours"
],
"title": "gpufeed_estimate_trainingArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "gpufeed_estimate_trainingDictOutput"
}⚪gpufeed_recommend(gpu, hours, count)
Return a bounded lowest-observed-cost recommendation for further verification.
输入模式
{
"type": "object",
"properties": {
"gpu": {
"title": "Gpu",
"type": "string"
},
"hours": {
"default": 1,
"title": "Hours",
"type": "number"
},
"count": {
"default": 1,
"title": "Count",
"type": "integer"
}
},
"required": [
"gpu"
],
"title": "gpufeed_recommendArguments"
}输出模式
{
"type": "object",
"additionalProperties": true,
"title": "gpufeed_recommendDictOutput"
}社区
证据
最近观测
已验证未记录版本11 个工具
已验证未记录版本11 个工具