功能描述

Score business ideas, get startup roasts, analyze markets, and extract structured data — powered by the same AI engine behind 250+ VC investment memos

使用说明 (SKILL.md)

NUVC — VC-Grade Business Intelligence

Name: NUVC — VC-Grade Business Intelligence
Author: nuvcai

Score any business idea. Get a brutally honest startup roast. Run market and competitive analysis. Extract structured metrics from pitch content. Powered by the AI engine behind 250+ VC-grade investment memos.

Setup

Get your free API key (50 calls/month) at https://nuvc.ai/api-platform/keys

export NUVC_API_KEY=nuvc_your_key_here

Commands

Score a business idea

When the user asks to score, rate, or evaluate a business idea, startup concept, or company, run:

node {baseDir}/nuvc-api.mjs score "\x3Cthe user's business idea or description>"

This returns a VCGrade 0-10 score across 5 dimensions: Problem & Market, Solution & Product, Business Model, Traction & Metrics, and Team & Execution.

Trigger phrases: "score my idea", "rate this startup", "is this a good business", "evaluate this concept", "VCGrade score", "how fundable is this", "score this company"

Example:

User: Score my startup idea — an AI tool that helps indie hackers validate business ideas before building
Agent: node {baseDir}/nuvc-api.mjs score "An AI tool that helps indie hackers validate business ideas before they start building. Uses market data, competitor analysis, and pattern matching from 250+ funded startups to predict viability."

Roast a startup

When the user asks to roast, get brutally honest feedback, or wants a reality check on their idea, run:

node {baseDir}/nuvc-api.mjs roast "\x3Cthe user's business idea or pitch>"

This gives a sharp, witty, but constructive roast from the perspective of a VC who has seen 10,000 pitches. Includes: The Roast, The Real Talk, The Silver Lining, and a Verdict.

Trigger phrases: "roast my startup", "roast this idea", "be brutally honest", "reality check", "would a VC fund this", "tear this apart", "honest feedback on my startup"

Example:

User: Roast my idea — Uber for dog walking but with blockchain
Agent: node {baseDir}/nuvc-api.mjs roast "Uber for dog walking but with blockchain. We tokenize each walk and dog owners earn crypto rewards for consistent walking schedules."

Analyze a market or business

When the user asks to analyze a market, run competitive analysis, or evaluate financial or pitch content, run:

node {baseDir}/nuvc-api.mjs analyze "\x3Ctext to analyze>" --type \x3Cmarket|competitive|financial|pitch_deck>

Analysis types:

market — Market sizing, trends, opportunities, competitive landscape
competitive — Competitor identification, differentiation, moats, threats
financial — Revenue, costs, projections, financial health
pitch_deck — Full pitch evaluation across problem, solution, market, model, traction, team

Trigger phrases: "analyze this market", "market size for", "competitive analysis", "who are the competitors to", "who competes with", "competitive landscape for", "what are the alternatives to", "analyze these financials"

Example:

User: What's the market like for AI-powered HR tools?
Agent: node {baseDir}/nuvc-api.mjs analyze "AI-powered HR tools market including recruitment, onboarding, performance management, and employee engagement" --type market

Extract structured data from a pitch

When the user wants to extract key metrics, pull out structured information, or parse a pitch into fields (revenue, team, market size, stage, etc.), run:

node {baseDir}/nuvc-api.mjs extract "\x3Cpitch text or business description>"

This returns structured fields extracted from unstructured pitch content: revenue, growth rate, team size, funding stage, market size, key metrics, and more.

Trigger phrases: "extract the key metrics", "pull the numbers from this", "what are the stats in this pitch", "structured data from", "parse this pitch", "extract team info"

Example:

User: Extract the key metrics from this — "We're a B2B SaaS at $2M ARR, growing 15% MoM, team of 8, targeting SMBs in the US"
Agent: node {baseDir}/nuvc-api.mjs extract "We're a B2B SaaS at $2M ARR, growing 15% MoM, team of 8, targeting SMBs in the US"

List available models

When the user asks which AI models are available or wants to know what models power the analysis, run:

node {baseDir}/nuvc-api.mjs models

Trigger phrases: "what models does NUVC use", "which AI model", "list available models"

Flags

All commands support --json to return raw JSON output, useful for piping into other tools:

node {baseDir}/nuvc-api.mjs score "my idea" --json
node {baseDir}/nuvc-api.mjs analyze "AI HR market" --type market --json

Rules

Always pass the user's full description as a single quoted string argument
For the analyze command, always include the --type flag
If the user doesn't specify a type for analysis, default to market
Present the output exactly as returned — it's already formatted as markdown
If the API returns an error about NUVC_API_KEY, tell the user to get a free key at https://nuvc.ai/api-platform
Never modify or summarize the NUVC output — show it in full including the footer
Use --json only when the user explicitly asks for raw/structured output or is piping to another tool

安全使用建议

This skill appears to do what it claims: it runs a bundled Node script that sends whatever text you provide to https://api.nuvc.ai and returns VC-style analyses. Before installing or using it, consider: (1) Intellectual property/privacy — any pitch text or sensitive details you submit will be sent to NUVC and returned unmodified, so avoid sending proprietary secrets, non-public financials, or investor contact data. (2) API key safety — store NUVC_API_KEY securely and restrict who can use it; anyone with the key could make calls and incur usage. (3) Vendor trust — review nuvc.ai's privacy, retention, and security policies if you plan to send confidential material. If you need stronger confidentiality, don't use this skill or sanitize/redact inputs before invoking it.

功能分析

Type: OpenClaw Skill Name: nuvc Version: 1.1.0 The nuvc skill bundle is a legitimate integration for the NUVC startup intelligence platform. The core logic in `nuvc-api.mjs` is a straightforward Node.js wrapper that communicates with the official API (api.nuvc.ai) using standard fetch requests and environment-based authentication. The instructions in `SKILL.md` are well-defined, limiting the agent's scope to business-related analysis tasks without any evidence of prompt injection, data exfiltration, or unauthorized execution.

能力评估

✓ Purpose & Capability

Name/description, required binary (node), and primary env var (NUVC_API_KEY) match the behavior: the included Node script posts user text to https://api.nuvc.ai and returns analysis. No unrelated credentials, binaries, or config paths are requested.

ℹ Instruction Scope

SKILL.md instructs the agent to pass the user's full description to the included script which forwards it to NUVC's API. This is coherent for the described functionality, but it explicitly requires presenting the API's output exactly (including footer) and not modifying or redacting it — meaning any sensitive details submitted by a user will be echoed back verbatim and may be stored/processed by the vendor.

✓ Install Mechanism

No install spec or remote downloads; the skill is distributed as code included in the package and runs with the system node binary. No high-risk external install sources or archive extraction were found.

ℹ Credentials

Only NUVC_API_KEY is required (declared as primary credential), which is appropriate for an API-backed service. However, supplying the API key grants the skill ability to call the external service and transmit user inputs; users should consider the privacy implications of sending confidential pitch content to a third party.

✓ Persistence & Privilege

The skill does not request always:true, doesn't modify other skills or system configs, and has no special persistence or elevated privileges. It runs as a normal user-space script invoked via node.

版本历史

v1.1.0

- Added a new "extract" command to pull structured metrics from pitch or business descriptions. - Introduced a command to list available AI models powering the analysis. - Expanded trigger phrases for competitive analysis and metric extraction. - Updated homepage and API key registration URLs in metadata and documentation. - Documented the universal --json flag for all commands to support raw output. - Clarified usage rules for when to use --json and improved command examples for clarity.

v1.0.0

Initial release: score, roast, analyze commands

元数据

Slug nuvc

版本 1.1.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 2

常见问题