Vela Breakout
What's growing in tech, measured daily: AI apps, models, dev tools, software spend, hiring.
我該用這個嗎
品質與安全性
根據工具定義與協定合規性的自動化分析。
上下文成本
這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高，可用於其他工作的注意力就越少。
安裝
一鍵安裝
將以下內容加入你的 `claude_desktop_config.json` 檔案：
{
"mcpServers": {
"breakout": {
"url": "https://breakout.vela.partners/mcp"
}
}
}遠端端點
https://breakout.vela.partners/mcpstreamable-http它能做什麼
工具清單
工具（6）
⚪ask(question)
Answer any question about what is growing in tech with checked figures and links to the charts. Best first call. Shows the chart inline where the client supports it.
輸入結構描述
{
"type": "object",
"properties": {
"question": {
"type": "string",
"minLength": 3,
"maxLength": 300,
"description": "The question in plain words"
}
},
"required": [
"question"
],
"$schema": "http://json-schema.org/draft-07/schema#"
}🟡find_breakouts(group, window_days, limit, min_size, steady)
The fastest-growing items of one group over a window. Groups: ai_apps (AI apps: tokens); ai_models (AI models: tokens a week); ai_labs (AI labs: share of all AI tokens); ai_work (AI work: open jobs); open_models (Open models: downloads in 30 days); editor_extensions (Code editor extensions: installs); open_editor_extensions (Open-source editor extensions: installs); workplace_add_ons (Workplace add-ons: installs); browser_extensions (Browser extensions: installs); agent_skills (Agent skills: installs a week); agent_tools (Agent tools: downloads a week); company_packages (Company packages: downloads in 30 days); business_software (Business software: adoption score); web_technologies (Web technologies: websites); company_hiring (Companies hiring: open roles); company_buzz (In the news: points in 12 months).
輸入結構描述
{
"type": "object",
"properties": {
"group": {
"type": "string",
"enum": [
"ai_apps",
"ai_models",
"ai_labs",
"ai_work",
"open_models",
"editor_extensions",
"open_editor_extensions",
"workplace_add_ons",
"browser_extensions",
"agent_skills",
"agent_tools",
"company_packages",
"business_software",
"web_technologies",
"company_hiring",
"company_buzz"
],
"description": "Group id from list_groups"
},
"window_days": {
"description": "Leave out for the group's usual window",
"anyOf": [
{
"type": "number",
"const": 7
},
{
"type": "number",
"const": 30
},
{
"type": "number",
"const": 90
},
{
"type": "number",
"const": 365
}
]
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 25
},
"min_size": {
"description": "Leave out tiny items (default median)",
"type": "string",
"enum": [
"median",
"p75",
"none"
]
},
"steady": {
"description": "Set true for 'steadily', 'starting to', 'sustained': keeps only items also up over a longer window",
"type": "boolean"
}
},
"required": [
"group"
],
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢get_company(name_or_domain)
Everything measured about one company: each signal's growth, how it compares with peers, and its page.
輸入結構描述
{
"type": "object",
"properties": {
"name_or_domain": {
"type": "string",
"minLength": 2,
"maxLength": 120,
"description": "Company name or web domain, e.g. supabase.com"
}
},
"required": [
"name_or_domain"
],
"$schema": "http://json-schema.org/draft-07/schema#"
}🟡research_ideas(direction, limit)
AI research ideas by change in their share of all papers, last 3 months against the 3 before.
輸入結構描述
{
"type": "object",
"properties": {
"direction": {
"type": "string",
"enum": [
"rising",
"fading"
]
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 25
}
},
"required": [
"direction"
],
"$schema": "http://json-schema.org/draft-07/schema#"
}⚪rising_companies(limit)
Companies breaking out on two or more measures at once, ranked by how much.
輸入結構描述
{
"type": "object",
"properties": {
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 25
}
},
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢list_groups
Every group you can ask about, with its unit, caveat and usual window.
輸入結構描述
{
"type": "object",
"properties": {},
"$schema": "http://json-schema.org/draft-07/schema#"
}社群
證據