Signalcrest
Cross-source trend signals from 10 developer communities, scored every 30 minutes.
我该使用它吗
质量与安全性
发现（2）
- HIGHTool poisoning patterns detected
- INFOTool description contains placeholder or incomplete text在 list_signals 中
基于对工具定义和协议合规性的自动分析。
上下文开销
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": {
"signalcrest": {
"url": "https://www.signalcrest.app/api/mcp"
}
}
}远程端点
https://www.signalcrest.app/api/mcpstreamable-http它能做什么
工具清单
工具（4）
🟢list_signals(category, limit)
Top technology signals ranked by momentum right now, scored every 30 minutes across ten developer and builder communities (Hacker News, GitHub, Stack Overflow, npm, Hugging Face, Dev.to, Lobsters, Product Hunt, arXiv, crypto governance forums). Call this to find what is gaining traction before it reaches search-trend tools — e.g. when the user asks what is emerging, trending, or worth paying attention to in a technology area. Returns live unredacted data with no delay.
输入模式
{
"type": "object",
"properties": {
"category": {
"type": "string",
"enum": [
"ai",
"devtools",
"security",
"web",
"data",
"crypto",
"hardware",
"business",
"science",
"other"
],
"description": "Restrict to one category. Omit for all categories."
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 200,
"description": "How many signals to return (default 50, max 200)."
}
},
"required": [],
"additionalProperties": false
}🟢get_topic(id)
Full detail and complete score history for one topic, by id (e.g. "hn:48910545", "gh:1287845516"). Call this after list_signals to see how a specific signal has moved over time, its cross-source corroboration, and its lifecycle stage. Ids come from list_signals.
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Topic id as returned by list_signals, e.g. \"hn:48910545\"."
}
},
"required": [
"id"
],
"additionalProperties": false
}🟢list_entities(category, limit)
Trending entities (products, companies, libraries, technologies) ranked by cross-source momentum. An entity aggregates every post, repo and release mentioning it across ten developer and builder communities — one row per trend rather than per post. Prefer this over list_signals when the user asks about trends or technologies; use list_signals for individual posts.
输入模式
{
"type": "object",
"properties": {
"category": {
"type": "string",
"enum": [
"ai",
"devtools",
"security",
"web",
"data",
"crypto",
"hardware",
"business",
"science",
"other"
],
"description": "Restrict to one category. Omit for all categories."
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 200,
"description": "How many entities to return (default 50, max 200)."
}
},
"required": [],
"additionalProperties": false
}🟢get_entity(name)
Full detail for one entity by name (e.g. "kubernetes"): current composite score, complete score history, which communities are talking about it, and the live member items (evidence) behind the score. Names come from list_entities.
输入模式
{
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "Entity name as returned by list_entities, e.g. \"kubernetes\"."
}
},
"required": [
"name"
],
"additionalProperties": false
}社区
证据