Real Alpha
What AI founders and researchers said in long interviews: quotes, speakers, dates and links to cite.
我该使用它吗
质量与安全性
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": {
"moments": {
"url": "https://realalpha.app/mcp"
}
}
}远程端点
https://realalpha.app/mcpstreamable-http它能做什么
工具清单
工具(6)
🟢search_moments(query, person, company, show, since, ...)
Search the moments Real Alpha cut from long interviews with the people building AI. Matches the words of `query` against what each moment is about, the speaker's quote and its takeaways, best match first. Filter by who is talking (`person`), a company (someone from it talking, or it being discussed), a show, or how recent. Each result has the speaker, what they do, the show, the date, their quote, and the moment's page to cite.
输入模式
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"maxLength": 200,
"description": "What to look for, in a few words: \"AI bubble\", \"coding agents replacing engineers\", \"scaling laws\""
},
"person": {
"description": "Only moments where this person is talking, by name: \"Sam Altman\", \"Dario Amodei\"",
"type": "string",
"maxLength": 100
},
"company": {
"description": "Only moments from someone at this company, or about it: \"Anthropic\", \"Nvidia\"",
"type": "string",
"maxLength": 100
},
"show": {
"description": "Only moments from this show: \"Lenny's Podcast\", \"Dwarkesh Podcast\", \"No Priors\"",
"type": "string",
"maxLength": 100
},
"since": {
"description": "Only episodes out on or after this date, YYYY-MM-DD",
"type": "string",
"pattern": "^\\d{4}-\\d{2}-\\d{2}$"
},
"limit": {
"description": "How many moments to return, 8 if not given, 20 at most",
"type": "integer",
"minimum": 1,
"maximum": 20
}
},
"required": [
"query"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_moment(slug)
One moment in full, by its slug (from search results) or its realalpha.app/s/ link: the speaker and what they do, the show and date, their quote, the summary and takeaways, its topics, the big questions it answers and how, and the episode's other best minutes.
输入模式
{
"type": "object",
"properties": {
"slug": {
"type": "string",
"minLength": 1,
"maxLength": 200,
"description": "The moment's slug, like \"ai-is-widening-the-gap-between-startups-and-incumbents\", or its page's link"
}
},
"required": [
"slug"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢what_did_they_say(person, topic, limit)
What one person said in the interviews Real Alpha cut, newest first: every moment where they are talking, or only those on a topic, best match first. Also says who they are and links their page on Real Alpha, which lists all of it.
输入模式
{
"type": "object",
"properties": {
"person": {
"type": "string",
"minLength": 2,
"maxLength": 100,
"description": "Their name: \"Jensen Huang\", \"Demis Hassabis\", or a surname when only one person has it"
},
"topic": {
"description": "A topic or a few words: one of Real Alpha's topics (Frontier models, Open-weight models, Reasoning & RL, Scaling laws, AGI & superintelligence, Interpretability, Evals & benchmarks, Training data, World models, AI agents, Coding agents, Building AI products, Enterprise AI, Consumer AI, Voice AI, Image & video models, Data & databases, Chips & GPUs, Data centers & energy, Inference & cost, Robotics, Self-driving, Science & math, Biology & drugs, Health & medicine, Safety & alignment, Policy & regulation, US–China race, Cybersecurity, Jobs & the economy, AI startups, Investing in AI, Big Tech's AI bets, AI business models, Talent & hiring) or anything else, like \"open source\"",
"type": "string",
"maxLength": 100
},
"limit": {
"description": "How many moments to return, 10 if not given, 20 at most",
"type": "integer",
"minimum": 1,
"maximum": 20
}
},
"required": [
"person"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢answers_to(question)
How the people building AI answer one of the big questions people ask about it, from the moments Real Alpha cut: grouped by which way they answer, the biggest group first, each with what the speaker says, who they are, the show and date. The questions: is-ai-a-bubble (Is AI a bubble?); which-jobs-will-ai-replace (Which jobs will AI replace, and which are safe?); will-ai-replace-software-engineers (Will AI replace software engineers?); will-ai-replace-doctors (Will AI replace doctors?); will-ai-replace-lawyers (Will AI replace lawyers?); is-saas-dead (Is SaaS dead?); when-will-we-get-agi (When will we get AGI?); is-openai-profitable (How does OpenAI make money, and is it profitable?); ai-predictions (What will AI look like in the next few years?); are-ai-agents-ready (Are AI agents ready for real work?); is-ai-paying-off (Is AI paying off for companies?); is-ai-dangerous (Is AI dangerous?); will-ai-make-cyberattacks-worse (Will AI make cyberattacks worse?); can-anyone-compete-with-nvidia (Can anyone compete with Nvidia?); is-there-a-gpu-shortage (Is there still a GPU shortage?); is-there-enough-power-for-ai (Is there enough power for AI?); is-moores-law-dead (Is Moore’s law dead?); is-ai-progress-slowing (Is AI progress slowing down?); is-open-source-ai-catching-up (Is open-source AI catching up?); is-china-catching-up-in-ai (Is China catching up in AI?); is-ai-overhyped (Is AI overhyped?); will-ai-replace-google-search (Will AI replace Google search?); is-vibe-coding-the-future (Is vibe coding the future?); is-a-computer-science-degree-worth-it (Is a computer science degree still worth it?); are-ai-wrappers-a-real-business (Are AI wrappers a real business?); how-do-ai-companies-make-money (How do AI companies make money?); should-ai-be-regulated (Should AI be regulated?); will-ai-cure-diseases (Will AI cure diseases?); when-will-humanoid-robots-be-useful (When will humanoid robots be useful?); are-self-driving-cars-safe (Are self-driving cars safe?); will-ai-boost-economic-growth (Will AI boost economic growth?); will-ai-replace-accountants (Will AI replace accountants?); will-ai-replace-teachers (Will AI replace teachers?); will-ai-replace-customer-service (Will AI replace customer service jobs?).
输入模式
{
"type": "object",
"properties": {
"question": {
"type": "string",
"enum": [
"is-ai-a-bubble",
"which-jobs-will-ai-replace",
"will-ai-replace-software-engineers",
"will-ai-replace-doctors",
"will-ai-replace-lawyers",
"is-saas-dead",
"when-will-we-get-agi",
"is-openai-profitable",
"ai-predictions",
"are-ai-agents-ready",
"is-ai-paying-off",
"is-ai-dangerous",
"will-ai-make-cyberattacks-worse",
"can-anyone-compete-with-nvidia",
"is-there-a-gpu-shortage",
"is-there-enough-power-for-ai",
"is-moores-law-dead",
"is-ai-progress-slowing",
"is-open-source-ai-catching-up",
"is-china-catching-up-in-ai",
"is-ai-overhyped",
"will-ai-replace-google-search",
"is-vibe-coding-the-future",
"is-a-computer-science-degree-worth-it",
"are-ai-wrappers-a-real-business",
"how-do-ai-companies-make-money",
"should-ai-be-regulated",
"will-ai-cure-diseases",
"when-will-humanoid-robots-be-useful",
"are-self-driving-cars-safe",
"will-ai-boost-economic-growth",
"will-ai-replace-accountants",
"will-ai-replace-teachers",
"will-ai-replace-customer-service"
],
"description": "The question's id, like \"is-ai-a-bubble\""
}
},
"required": [
"question"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢transcript_excerpt(video, second)
About a minute of an episode's transcript, from a second into the video, for an episode Real Alpha cut: the lines with their times and, where the transcript says, who is talking, plus the link that plays the episode from there. Use it to check the exact words around a moment's quote (a moment's `episode_url` has the video id and second).
输入模式
{
"type": "object",
"properties": {
"video": {
"type": "string",
"minLength": 11,
"maxLength": 200,
"description": "The YouTube video id, like \"dQw4w9WgXcQ\", or a realalpha.app/v/ or YouTube link"
},
"second": {
"type": "number",
"minimum": 0,
"maximum": 86400,
"description": "Seconds into the video: 1910 for 31:50"
}
},
"required": [
"video",
"second"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢latest_moments(n, show)
The newest moments Real Alpha cut, newest episode first: each episode's best minutes, from every show or only one.
输入模式
{
"type": "object",
"properties": {
"n": {
"description": "How many, 10 if not given, 20 at most",
"type": "integer",
"minimum": 1,
"maximum": 20
},
"show": {
"description": "Only from this show: \"Lenny's Podcast\"",
"type": "string",
"maxLength": 100
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}社区
证据