惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Microsoft Azure Blog
Microsoft Azure Blog
Google Online Security Blog
Google Online Security Blog
D
Darknet – Hacking Tools, Hacker News & Cyber Security
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Simon Willison's Weblog
Simon Willison's Weblog
T
Threat Research - Cisco Blogs
C
CXSECURITY Database RSS Feed - CXSecurity.com
L
Lohrmann on Cybersecurity
www.infosecurity-magazine.com
www.infosecurity-magazine.com
Forbes - Security
Forbes - Security
P
Palo Alto Networks Blog
Schneier on Security
Schneier on Security
S
Schneier on Security
T
Tor Project blog
cs.CL updates on arXiv.org
cs.CL updates on arXiv.org
WordPress大学
WordPress大学
The Hacker News
The Hacker News
Hacker News - Newest:
Hacker News - Newest: "LLM"
罗磊的独立博客
Application and Cybersecurity Blog
Application and Cybersecurity Blog
F
Fortinet All Blogs
博客园 - 三生石上(FineUI控件)
小众软件
小众软件
C
Check Point Blog
Stack Overflow Blog
Stack Overflow Blog
Blog — PlanetScale
Blog — PlanetScale
雷峰网
雷峰网
S
Security @ Cisco Blogs
PCI Perspectives
PCI Perspectives
Spread Privacy
Spread Privacy
W
WeLiveSecurity
SecWiki News
SecWiki News
A
About on SuperTechFans
H
Help Net Security
博客园 - 司徒正美
Recent Commits to openclaw:main
Recent Commits to openclaw:main
爱范儿
爱范儿
S
Securelist
M
MIT News - Artificial intelligence
云风的 BLOG
云风的 BLOG
月光博客
月光博客
Jina AI
Jina AI
博客园 - 叶小钗
Vercel News
Vercel News
阮一峰的网络日志
阮一峰的网络日志
Recent Announcements
Recent Announcements
S
Secure Thoughts
The Cloudflare Blog
美团技术团队
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor GitHub - GenAI-Gurus/awesome-eu-ai-act: Curated tools, official sources, OSS, templates, and guides for EU AI Act compliance. Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge Can AI be a 'child of God'? Inside Anthropic's meeting with Christian leaders How to Switch AI Chatbots and Why You Might Want To GitHub - MattMessinger1/agentic_refund_guardrail: Safe refund policy layer for AI agents — Python + TypeScript. Same behavior, shared tests. Adam/papers/emergent_values_whitepaper.md at master · strangeadvancedmarketing/Adam Ask HN: How do you stop playing 20 questions with your AI coding tools How far can automation and AI support psychotherapy? - @theU GitHub - stagas/rtdiff: realtime git diff gui and AI-assisted commits A Mac Studio for Local AI — 6 Months Later A History of the Early Years of AI at the University of Edinburgh Why AI Coding Tools Still Feel Stuck on Localhost MSN AI Datacenters Are Becoming Strategic Targets twitter.com Penn Researchers Use AI to Surface Unreported GLP-1 Side Effects in Reddit Posts Show HN: MoodSense AI (ML and FastAPI and Gradio, Deployed on Hugging Face) Moodsense Ai - a Hugging Face Space by aman179102 AI models are terrible at betting on soccer—especially xAI Grok GitHub - xialeistudio/echoic GitHub - HimashaHerath/github-dev-wrapped: AI-powered weekly GitHub activity reports deployed to GitHub Pages GitHub - alejandrobalderas/claude-code-from-source: Architecture, patterns & internals of Anthropic's AI coding agent — reverse-engineered from source maps AI and Tech brief: Ireland ascendant GitHub - Titovilal/context0: Context0 - Never Surrender Training for a Marathon with an AI Coach: What Worked and What Didn't Cyber Pulse: Agentic Intel - Apps on Google Play I Built an AI PR Reviewer That Catches Bugs by Not Looking for Bugs Gen Z workers are so fearful AI will take their job they’re intentionally sabotaging their company’s AI rollout | Fortune How AI Is Reimagining the Game of Golf–For Both Players and Courses GitHub - nattergabriel/reseed: A CLI tool for managing and distributing agent skills across projects Is SVG the final frontier? My AI workflow evolved from prompts to a near-autonomous workflow MLSharp Help - 3DGS Viewer & Generator I put my cognitive field based AI's runtime on GitHub Is Numble the first AI-proof game? A3: Kubernetes for autonomous AI agent fleets | Emergent Principles Deepali Vyas ("The Elite Recruiter") GitHub - msmarkgu/RelayFreeLLM: A restful API designed to route user prompts to various AI model providers. Unionized ProPublica staff are on strike over AI, layoffs, and wages Unleashing the Advantage of Quantum AI We're heading for an AI-fueled 'dementia crisis,' brain scientist warns The AI-Assisted Breach of Mexico's Government Infrastructure [pdf] GitHub - stef41/lmscan: 🔍 Detect AI-generated text and fingerprint which LLM wrote it. Open-source GPTZero alternative. Zero dependencies, works offline. MSN GitHub - visionscaper/collabmem: Enabling long-term collaboration with Agentic AI - building up episodic and world model memory over time with in-context awareness We gave an AI a 3 year retail lease in SF and asked it to make a profit | Andon Labs AI Code is Hollowing Out Open Source, and Maintainers are Looking the Other Way What leaked "SteamGPT" files could mean for the PC gaming platform's use of AI AI is the boss at this retail store. What could go wrong? GitHub - Wuzu11517/agentic-proxy: Local proxy meant to help reduce With Drones, Geophysics and ArtificiaI Intelligence, Researchers Prepare to Do Battle Against Land Mines A Single Operator, Two AI Platforms, Nine Government Agencies: The Full Technical Report 在 Steam 上购买 FriedrichAI: Offline AI 立省 10% GitHub - inevolin/resume-cli: Hit Claude usage limits? Resume any AI coding session elsewhere. Switch tools at zero friction. GitHub - atripati/ark: AI Runtime Kernel — a context operating system for AI agents. Eliminates tool bloat, loads only what’s needed, and gives LLMs their reasoning space back. How to Build a Secure AI PR Reviewer with Claude, GitHub Actions, and JavaScript This Startup Wants You to Pay Up to Talk With AI Versions of Human Experts Intel Arc Pro B70 Brings 32GB VRAM to Local AI for $949 WordPress 7.0: The Good, the AI, and the Still Missing AI on the couch: Anthropic gives Claude 20 hours of psychiatry IatroBench: Pre-Registered Evidence of Iatrogenic Harm from AI Safety Measures AI Agents Know About Supabase. They Don't Always Use It Right. The history and future of AI at Google, with Sundar Pichai Inside an AI‑enabled device code phishing campaign How Meta Used AI to Map Tribal Knowledge in Large-Scale Data Pipelines AI for Systems: Using LLMs to Optimize Database Query Execution Forecasting the Economic Effects of AI Introducing Tinker: Play with AI, bring your ideas to life AI sheds light on an ancient gaming mystery People really hate AI but not as much as Iran—or Democrats | Fortune What is an AI Product Engineer? Phoebe Gates wants her $185 million AI startup to succeed with 'no ties to my privilege or my last name': 'I have a chip on my shoulder' | Fortune
GitHub - maxoliverbr/vcupid-plugin
maxoliverbr · 2026-06-26 · via Hacker News - Newest: "AI"

vcupid-logo

VCupid Skills — AI Fundraising Toolkit for Founders

A set of Skills that turn your startup profile into a full VC fundraising workflow: pipeline research, fund matching, partner profiling, outreach drafting, meeting prep, post-meeting debrief, term sheet analysis, and strategy.


Full Fundraising Workflow

Run commands in this order for a complete fundraising campaign:

# 0. Create your STARTUP_PROFILE.md
# Use the example file with your information. 

# 1. Build the strategy
/vcraise

# 2. Research the landscape
/vclist

# 3. For each Tier 1 fund — legitimacy check first:
/vcposer <fund name>              # Is this fund real and active? Drop if score < 40.

# 4. For each fund that passes vcposer (score 60+):
/vcmatch <fund name>              # Does their mandate fit your startup?
/vcperks <fund name>              # What do they offer beyond the check?

# 5. Stress-test your pitch against the funds you're pursuing:
/vcdevil                          # 10 lethal questions — know what to defend
/vcangel                          # 10 strongest signals — know what to lead with

# 6. For each fund with a vcmatch Pursue or Warm Up verdict:
/vcpartner <fund> <partner name>
/vcintro vcmatch-<fund>.md        # send outreach
/vclp vcmatch-<fund>.md           # attach the one-pager

# 7. When a meeting is booked:
/vcprep vcmatch-<fund>.md

# 8. After the meeting:
/vcdebrief vcmatch-<fund>.md        # signal read, follow-up email, next step

# 9. When you receive a term sheet:
/vcterm termsheet-<fund>.md         # clause analysis, score, negotiation priority list

# Anytime — review your full pipeline:
/vctrack

File naming convention:

File Purpose
STARTUP_PROFILE.md Your startup data — update as traction grows
vclist.md Master fund pipeline
vcposer-<fund>.md Poser check — fund vitality, thesis authenticity, lead capacity
vcmatch-<fund>.md Deep analysis per fund
vcperks-<fund>.md Perks breakdown — value beyond capital, scored by relevance
vcpartner-<fund>-<name>.md Partner brief
vcintro-<fund>.md Outreach drafts
vclp-<fund>.md Fund-specific one-pager
vcprep-<fund>.md Meeting prep
vcdebrief-<fund>.md Post-meeting debrief — signals, follow-up email, next step
vcterm-<fund>.md Term sheet analysis — clause scores, negotiation priority list
vctrack.md Pipeline dashboard — status of all funds at a glance
vcraise.md Fundraising strategy
vcdevil.md Adversarial stress-test — 10 lethal questions
vcangel.md Champion's brief — 10 strongest signals to lead with

Installation

These commands are available as the vcupid Claude Code plugin. Install with a single command — works in Claude Code web or CLI:

curl -fsSL https://raw.githubusercontent.com/maxoliverbr/vcupid-plugin/main/install.sh | bash

The script clones the plugin to ~/.claude/plugins/vcupid/, registers it in ~/.claude/plugins/installed_plugins.json, and enables it in ~/.claude/settings.json. It is idempotent — safe to re-run to update.

Restart Claude Code. All /vc* commands will be available in any directory that contains a STARTUP_PROFILE.md.

Local development install

If you've cloned the repo locally, you can install from the clone directly:

cd ~/dev/vcupid-plugin && bash install.sh

This registers the local clone path so edits to skill files take effect immediately without re-running the installer.

Manual installation

Register in ~/.claude/plugins/installed_plugins.json:

"vcupid@local": [{
  "scope": "user",
  "installPath": "/home/<you>/dev/vcupid",
  "version": "1.0.0",
  "installedAt": "<ISO timestamp>",
  "lastUpdated": "<ISO timestamp>"
}]

Enable in ~/.claude/settings.json:

"enabledPlugins": {
  "vcupid@local": true
}

Tips

  • Keep STARTUP_PROFILE.md current. Every new traction milestone (signed LOI, new partnership, first revenue) changes what funds you can approach and what you can say. Update it and re-run /vcraise monthly.
  • Run /vcmatch before spending time on outreach. A 40/100 fit score is a no-go — don't write the email.
  • Use Variant B (/vcintro) over cold email whenever a warm path exists. A forwarded intro converts at 5–10x the rate of cold outreach.
  • The vcprep is for rehearsal, not for the room. Print it, rehearse it, then leave it behind — the meeting is a conversation, not a recital.
  • Run /vcdebrief the same day as the meeting. Memory fades fast and the follow-up email loses impact after 24 hours. Don't wait until the next morning.
  • Share /vcterm output with your attorney, not instead of one. The term sheet analysis tells you what to fight for — your lawyer tells you how to fight for it without blowing the deal.
  • Run /vctrack before every weekly review. If you're working 5+ funds in parallel, the pipeline table is the only way to see what's stale and what needs a nudge.
  • Want to add or change a skill? See Contributing below.

Prerequisites

All commands read STARTUP_PROFILE.md from your current working directory. Create this file before running any command.

Setting Up STARTUP_PROFILE.md

Create a markdown file in your project root with the following structure. The more detail you provide, the more specific and accurate every command output will be. Vague inputs produce generic outputs.

# Startup Profile: [Company Name]

## I. Company Overview
| Field | Detail |
|-------|--------|
| **Startup Name** | [Legal company name] |
| **One-Liner** | [One sentence: what you do and for whom] |
| **Core Problem** | [The specific problem you solve — include scale/urgency] |
| **Solution** | [What you built and how it works — be technical if relevant] |
| **Key Value Prop** | [The headline outcome you deliver: time saved, cost reduced, etc.] |
| **Category** | [e.g., Deep Tech / Infrastructure / Climate / SaaS] |
| **Location & Founded** | [City, State, Country. Month Year.] |
| **Website** | [URL] |

## II. Market & Business Model
| Field | Detail |
|-------|--------|
| **Target Customer** | [Specific buyer — not "enterprises", but "grid operators at electric cooperatives"] |
| **Market Opportunity** | [Beachhead TAM with source. Expansion path.] |
| **Business Model** | [How you charge: SaaS, value-based, usage, marketplace, etc.] |

## III. Traction & Milestones
| Field | Detail |
|-------|--------|
| **Product Stage** | [Idea / MVP / MVBP / Revenue / Growth] |
| **Product Launch** | [Target date or "Live as of [date]"] |
| **Traction** | [Be specific: design partners, LOIs, revenue, pilots, grants, gov partnerships] |
| **Referral** | [Any warm contacts who can open VC doors — Name, email, relationship] |

## IV. Team
| Role | Name | Key Highlights |
|------|------|---------------|
| **CEO** | [Name] | [3–5 most relevant credentials for investors] |
| **CTO** | [Name] | [3–5 most relevant credentials for investors] |
| [Other] | [Name] | [Credentials] |

## V. Fundraise
| Field | Detail |
|-------|--------|
| **Raise Amount** | [$XM] |
| **Instrument** | [SAFE / Priced Seed / TBD] |
| **Use of Funds** | [Top 3 line items] |
| **Milestone this raise funds** | [What you'll have accomplished when the money runs out] |

Tips for a strong profile:

  • Use specific numbers everywhere: "$5M DOE proposal", "38,000-member LinkedIn group", "$30.2M transmission project" — not "large DOE proposal"
  • Name your referrals with contact info — the outreach commands use them
  • Describe the problem with the urgency a VC would feel, not just the solution you built
  • List credentials that matter to investors, not your full resume

Commands

/vcraise — Fundraising Strategy Memo

Start here. Synthesizes your profile and any existing vcmatch/vclist files into a full strategy.

Reads: STARTUP_PROFILE.md + any vclist.md and vcmatch-*.md in the current directory
Saves: vcraise.md

Output sections:

  1. Current State — honest assessment of what you have and what's missing
  2. Raise Parameters — recommended amount, instrument, valuation cap, timeline
  3. Sequencing Strategy — who first, parallel vs. sequential, how to create urgency
  4. Milestone Map — which proof points unlock which tier of investors
  5. Month-by-Month Timeline — from today through close
  6. Risk Factors & Contingencies — what derails the raise and the fallback for each
  7. This Week — Next 3 Actions — concrete, specific, doable in 7 days

/vclist — VC Target Pipeline

Research the VC landscape and produce a ranked list of 15–20 funds to target.

Reads: STARTUP_PROFILE.md
Saves: vclist.md

Output: Three tiers of funds with rationale and suggested contact angle per fund:

  • Tier 1 — Pursue Now: Strong fit, correct stage, warm path exists
  • Tier 2 — Warm Up First: Good thesis fit, missing proof point or intro
  • Tier 3 — Monitor: Stage mismatch or unclear path — revisit later

Run this before /vcmatch to know which funds are worth the deep analysis.


/vcposer — VC Poser Detector

Pre-filter funds from your pipeline before committing to a full match analysis. Runs 8 evidence-based checks on whether a fund is actually active and writing checks — independent of whether their thesis fits your startup.

Example: /vcposer "Slow Ventures" or /vcposer a16z
Reads: STARTUP_PROFILE.md (for stage/sector calibration)
Saves: vcposer-<fund>.md

Output — Poser Score (0–100) across 8 checks:

  • Fund Vitality — Last investment date. >18 months silent = zombie risk.
  • Fund Health — Active fund with dry powder? Key partner departures?
  • Thesis Authenticity — Does their portfolio actually match their stated sector thesis?
  • Lead vs. Follow — Evidence of leading rounds, not just co-investing.
  • Stage Honesty — Stated stage vs. actual entry stage of recent investments.
  • Check Size Reality — Stated range vs. observable deal sizes.
  • Signal-to-Check Ratio — Conference/content volume vs. actual new deals.
  • Founder Sentiment — Public feedback patterns — ghosting, slow decisions, reneging.

Verdict tiers:

  • Legit (80–100): Active fund, real checks. Proceed to /vcmatch.
  • Probable (60–79): Yellow flags — prioritize if fit is strong, clarify open questions.
  • Watch List (40–59): Material poser signals. De-prioritize unless a warm intro exists.
  • Likely Poser (0–39): Don't waste cycles. Move to the next fund.

vcposer ≠ vcmatch. Poser Score answers "is this fund real and active?" Vcmatch answers "does their mandate fit our startup?" You need both. A fund can be fully legit but wrong for your stage — and a fund can claim a perfect thesis but be a zombie. Run /vcposer first, then /vcmatch only for funds that score 60+.


/vcmatch — Fund Fit Analysis

Deep research on a single fund against your startup profile. Spawns 4 parallel domain research sub-agents, synthesizes the results into a structured report, and scores the match across 8 dimensions summing to /100.

Example: /vcmatch a16z or /vcmatch "Lowercarbon Capital"
Reads: STARTUP_PROFILE.md
Saves: vcmatch-<fund>.md

How it works: Four sub-agents run in parallel — Thesis & Mandate, Portfolio Intelligence, People & Network, and Deal & Process — each with domain-specific search queries. The main agent synthesizes all four returns and scores the fit.

Output sections:

  1. Firm Snapshot — founded, model, key GPs, standard deal, fund size, notable portfolio
  2. Investment Thesis & Current Focus — table with Theme / Signal / Verbatim Quote
  3. Portfolio Pattern Analysis — what the last 24 months of investments reveal about actual deployment priorities
  4. Key People — partner roster with background, sector focus, and access dynamics
  5. Fit Assessment — 8-dimension score table summing to /100
  6. Why It Works — positive signals with named portfolio analogs
  7. Key Risks & Objections — material mismatches with specific math
  8. Deal Process Intel — decision timeline, diligence style, known pass triggers
  9. Best Approach — intro strategy, lead-with bullets, avoid list
  10. Warm Intro Path — named intermediaries and specific actions, ranked by likelihood
  11. Diligence Questions to Prepare — 5–8 fund-specific questions
  12. Cold Email — subject + 4-sentence body with fund-specific hook
  13. LinkedIn Message — ≤150 chars
  14. Gaps / Things to Verify
  15. Sources
  16. Verdict — table including Recommended Action

Fit Score thresholds: 75–100 → Pursue · 50–74 → Warm Up First · 0–49 → No-Go

Run /vcmatch before /vcpartner, /vcintro, /vclp, or /vcprep.


/vcperks — Fund Perks Researcher

Understand what a fund delivers beyond the check before you sign. Researches the full value a fund delivers beyond the check — credits and services, but also brand signal, strategic expertise, media reach, policy access, and follow-on capital pathway — scored by relevance to your startup.

Example: /vcperks a16z or /vcperks "First Round Capital"
Reads: STARTUP_PROFILE.md (for stage/sector calibration)
Saves: vcperks-<fund>.md

Part A — Financial Value (dollar estimates, [Confirmed]/[Reported] labels):

  • Cloud & Infrastructure — AWS/GCP/Azure credits, API platform discounts
  • Legal & Finance — discounted counsel, accounting, cap table tools (Carta, Pulley)
  • Talent & Recruiting — job boards, ATS discounts, executive recruiting relationships
  • Go-to-Market — PR/comms support, CRM and sales tool discounts
  • Network & Introductions — portfolio co. intros, LP relationships, co-investor paths
  • Operational Support — office space, co-working, founder summits
  • Formal Programs — structured accelerator tracks, office hours, mentorship

Part B — Non-Financial Value (impact ratings, qualitative assessment):

  • Brand & Signal Value — what the fund's name does for customer credibility, talent attraction, and future round momentum
  • Strategic Guidance & Pattern Recognition — named GP expertise, comparable investments they've led, operating partners
  • Content & Media Amplification — publishing access, newsletter reach, event speaking, social amplification
  • Policy & Regulatory Access — government relations team, compliance guidance, agency relationships
  • Follow-on Capital Pathway — pro rata appetite, reserve pool, bridge capacity, Series A syndication partners, fund lifecycle

Closes with a Gaps section (benchmarked against named peer funds) and a Questions to Ask the Fund Directly section from unresolved research.

For most seed-stage companies, the non-financial value — brand signal, strategic depth, follow-on pathway — is worth more over a 5-year relationship than the credits. /vcperks covers both.


/vcpartner — Partner Deep Profile

Know who's in the room before you walk in. Research a specific GP's personal thesis, deals they've led, public writing, and how they map to your startup.

/vcpartner <fund> <partner name>

Example: /vcpartner a16z "Erin Price-Wright"
Reads: STARTUP_PROFILE.md
Saves: vcpartner-<fund>-<lastname>.md

Output sections:

  1. Background — career path and pre-VC experience
  2. Personal Thesis — what they personally care about, with direct quotes
  3. Deals They've Led — companies, stage, sector, why they backed them
  4. Public Writing & Key Quotes — specific essays and clips relevant to your startup
  5. Personal Touchpoints — conferences, topics they post about, shared background
  6. Fit Assessment — how this partner specifically maps to your team and problem
  7. Conversation Starters — 3 specific, sourced things to bring up in the meeting

/vcintro — Outreach Drafter

Turn a vcmatch report into ready-to-send outreach. Three variants for every situation.

/vcintro <vcmatch-report.md>

Example: /vcintro vcmatch-a16z.md
Reads: STARTUP_PROFILE.md + vcmatch report
Saves: vcintro-<fund>.md

Output — three variants:

  • Variant A — Cold Email: 4-sentence direct email to the target partner. Subject line references their thesis.
  • Variant B — Warm Intro Request: A note to your referral contact (from your profile), ready to forward to the partner.
  • Variant C — LinkedIn DM: 3 sentences. No deck link. Opens with the fund-specific hook.

Each variant uses the verbatim opening hook from the vcmatch report — never generic. The skill recommends which variant to use based on your available warm paths.

⚠️ If the vcmatch Recommended Action is No-Go, this command will stop and tell you — don't send outreach to a no-go fund.


/vclp — Fund-Specific One-Pager

A one-page executive summary written entirely in the fund's language and framed around their thesis. Not a generic company description.

/vclp <vcmatch-report.md>

Example: /vclp vcmatch-a16z.md
Reads: STARTUP_PROFILE.md + vcmatch report
Saves: vclp-<fund>.md

Output sections (≤500 words total):

  • Headline — one-liner reframed in the fund's thesis language
  • The Problem — uses the fund's own published framing
  • The Solution — mapped to their portfolio language
  • Traction — only the proof points that scored highest for this fund
  • Market — sized in terms this fund cares about
  • Team — only the credentials that matter to this fund
  • The Ask — amount + milestone unlocked + timeline
  • Why Now — fund-specific urgency tied to their thesis

⚠️ Each one-pager is fund-specific. Do not send to other funds without running /vcmatch <fund> first.

Attach this to Variant A or B of /vcintro.


/vcprep — 15-Minute Pitch Meeting Prep

Full prep for a pitch meeting: timed agenda, verbatim talking points per segment, preemptive answers to every diligence question from the vcmatch report, red flag rebuttals, and the exact ask.

/vcprep <vcmatch-report.md>

Example: /vcprep vcmatch-a16z.md
Reads: STARTUP_PROFILE.md + vcmatch report
Saves: vcprep-<fund>.md

Output:

  • 15-minute timed agenda (8 segments)
  • Segment scripts: what to say, what to show, what NOT to say — for each segment
  • Preemptive answers to every diligence question from the vcmatch report (with "if they push" follow-ups)
  • Red flag rebuttals — one-liner responses for each vcmatch red flag
  • The Ask — exact amount, 90-day plan, leave-behind action
  • Meeting discipline — 3 rules for controlling the room

/vcdevil — The Startup Destroyer

Before you pitch anyone, get destroyed first. A cocky, arrogant VC partner who's seen a thousand pitches like yours tears your startup apart with the 10 hardest questions he can find in your profile.

Reads: STARTUP_PROFILE.md
Saves: vcdevil.md

Output: 10 lethal questions derived from your actual profile — not generic diligence, but the specific traps your startup walks into. Each question includes:

  • A setup line in character (bored, contemptuous, or mocking)
  • The question itself, phrased to sting
  • Why this kills you — the exact vulnerability it exposes
  • What a real answer looks like — the framework for a rebuttal that shuts him up

Run this before any outreach. The questions you can't answer in 30 seconds are the gaps to close first.


/vcangel — The Startup Champion

The other side of /vcdevil. A seasoned VC partner who champions deals in the partner meeting surfaces the 10 strongest signals in your profile — not cheerleading, but pattern recognition. The things that make skeptics go quiet and interested.

Reads: STARTUP_PROFILE.md
Saves: vcangel.md

Output: 10 insights derived from your actual profile — the specific assets most investors would overlook or underweight. Each insight includes:

  • A setup line in character (warm, perceptive, strategically excited)
  • The insight itself, phrased as the angel would say it in the partner meeting
  • Why this wins the room — the specific strength it names, with evidence from your profile
  • How to amplify this — a coaching prompt for making the point land harder

Run this alongside /vcdevil. Together they give you full-spectrum pitch prep: know what to defend, know what to lead with.


/vcdebrief — Post-Meeting Debrief

Turn your raw meeting memory into a structured debrief. Produces a signal read, an open diligence queue, a send-ready follow-up email, and a clear recommended next step.

/vcdebrief <vcmatch-report.md>

Example: /vcdebrief vcmatch-a16z.md
Reads: STARTUP_PROFILE.md + vcmatch report (+ asks you for meeting notes)
Saves: vcdebrief-<fund>.md

Output sections:

  1. Meeting Summary — attendees, duration, format
  2. Signal Read — green (interest), yellow (hesitation), red (pushback) signals from the meeting
  3. Diligence Prediction vs. Reality — which vcmatch questions came up, which didn't, what surprised you
  4. Open Diligence Queue — every item raised in the meeting with a draft answer and deadline
  5. Pitch Landing Assessment — what from the vcmatch angle resonated vs. missed
  6. Follow-Up Email — send-ready, subject line and body, references something specific from the meeting
  7. Recommended Next Step — one of: Request Second Meeting / Send Deck / Send Data Room / Give Space / Drop Fund
  8. Updated Fit Signal — revised read on interest level vs. the vcmatch prediction

Run this within 24 hours of every meeting. The follow-up email degrades sharply after that.


/vcterm — Term Sheet Analyzer

Benchmark every clause in a VC term sheet against current market norms, flag founder-unfriendly terms, and produce a ranked negotiation priority list with suggested pushback language.

/vcterm <termsheet-file.md>

Example: /vcterm termsheet-a16z.md
Reads: STARTUP_PROFILE.md + term sheet file
Saves: vcterm-<fund>.md

Output sections:

  1. Term Sheet at a Glance — instrument, amount, cap, discount, key dates
  2. Founder-Friendliness Score — /100 with breakdown (–15 per red flag, –5 per yellow flag)
  3. Clause Analysis Table — each term vs. market norm with 🟢🟡🔴 verdict
  4. Red Flag Deep Dives — real-world impact for each 🔴 clause
  5. Negotiation Priority List — ranked by founder harm, with suggested pushback language and likelihood of success
  6. Questions to Ask Before Signing — 5–8 diligence questions for the fund or your attorney

Verdict tiers:

  • 85–100 — Founder-Friendly: Standard or better. Sign with standard review.
  • 65–84 — Acceptable: Negotiate the reds.
  • 45–64 — Investor-Favorable: Push back on all reds, escalate yellows.
  • 0–44 — Aggressive: Get a second legal opinion before proceeding.

⚠️ This skill is informational only — it does not constitute legal advice. Share the output with your attorney before responding to the fund.


/vctrack — Pipeline Dashboard

Scan all vcupid output files in the current directory and generate a single status table showing every fund's stage, scores, and next recommended action.

No argument needed. Reads from the current working directory.
Saves: vctrack.md

Output:

  • Summary counts (total, active, no-go, stale)
  • Pipeline table: Fund | Poser Score | Fit Score | Recommended Action | Stage Reached | Next Step
  • Urgent Actions list — funds with missing follow-through flagged with the specific command to run

Run this any time to get a bird's-eye view of where your pipeline stands. Especially useful before investor update calls or weekly reviews.


Contributing

Contributions are welcome — bug fixes, clearer skill instructions, new commands, and documentation improvements.

Other platforms

We seek help testing the skills in other tools (Opencode, Codex, Cursor...) and llms.

Development setup

  1. Fork and clone the repo:
    git clone https://github.com/<your-username>/vcupid-plugin.git
    cd vcupid-plugin
  2. Install from your local clone (registers this path with Claude Code):
  3. Restart Claude Code. Skill changes in skills/ are picked up from the registered install path; you do not need to re-run the installer after every edit.

To validate skill structure (optional, requires skills-ref):

skills-ref validate skills/vcmatch/

install.sh runs the same validation automatically when skills-ref is on your PATH.

Repository layout

Path Purpose
skills/<name>/SKILL.md Skill definition — frontmatter, invocation, execution steps, output rules
skills/<name>/references/ Optional templates (e.g. output-format.md) linked from the skill
assets/ Shared assets (logo, STARTUP_PROFILE_TEMPLATE.md)
.claude-plugin/plugin.json Plugin metadata for Claude Code
install.sh Registers the plugin in ~/.claude/plugins/ and enables it in settings

Claude Code discovers skills from every skills/*/SKILL.md file. There is no separate manifest to update when you add a folder.

Editing an existing skill

  1. Open skills/<name>/SKILL.md.
  2. Keep the YAML frontmatter consistent with sibling skills (name, description, license, compatibility, allowed-tools, metadata).
  3. If the skill references an output template, update skills/<name>/references/output-format.md in the same change when the on-disk report format changes.
  4. Smoke-test in Claude Code: create a STARTUP_PROFILE.md in a test directory, run the /vc* command, and confirm the saved markdown matches the skill’s rules.

Adding a new skill

  1. Create skills/<command-name>/SKILL.md (command name should match the folder, e.g. skills/vcmatch//vcmatch).
  2. Follow the structure used by existing skills:
    • Frontmatter with a one-line description that includes usage (Usage: /vcfoo ...)
    • Invocation — syntax and examples
    • Execution Steps — ordered steps the agent must follow
    • Output Format / Output Rules — what to write and where to save (vcfoo-<fund>.md, etc.)
  3. Add optional references/output-format.md if the report has a fixed section layout.
  4. Document the command in this README under Commands and, if it fits the fundraising flow, in Full Fundraising Workflow.
  5. Run skills-ref validate skills/<command-name>/ if available.

Pull requests

  1. Branch from main with a focused change (one skill or one doc area per PR when possible).
  2. Describe what you changed and how you tested it (which /vc* command you ran, and whether output looked correct).
  3. Open a PR against main on GitHub. Issues and ideas are also welcome via GitHub Issues.

By contributing, you agree that your contributions are licensed under the same terms as the project (LICENSE).