Token导航 LogoToken导航

Agent Skill · 研究检索

saas-launch-audit(saas launch 审核)

用于辅助安全审计、权限检查、凭据风险、认证流程和常见漏洞排查。它适合让 Agent 梳理敏感配置、检查依赖风险、分析鉴权逻辑或生成安全复核清单。使用时不能把工具输出直接当最终结论,涉及密钥、令牌、用户数据或生产系统时,应先确认最小权限、脱敏方式和操作边界。

saas-launch-audit 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证为 MIT,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 46,029。

总安装 815周安装 35GitHub Stars 11下载 286核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:saas-launch-audit(saas launch 审核)
来源仓库:https://github.com/b-open-io/prompts
仓库路径:skills/saas-launch-audit
安装命令:
npx skills add https://github.com/b-open-io/prompts --skill saas-launch-audit
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/b-open-io/prompts --skill saas-launch-audit

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于辅助安全审计、权限检查、凭据风险和认证流程排查,适合 Agent 梳理敏感配置和鉴权逻辑。适用于生成安全复核清单、分析依赖风险和常见漏洞场景。通过 npx skills add 命令从指定 GitHub 仓库安装并使用。不能将工具输出直接作为最终结论,涉及密钥或生产系统时应确认最小权限和操作边界。

建议结合原始 README 和仓库内容进一步核验具体用法和功能边界。

来源仓库内路径:skills/saas-launch-audit

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/b-open-io/prompts --skill saas-launch-audit;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

saas-launch-audit 怎么安装?

在终端执行 npx skills add https://github.com/b-open-io/prompts --skill saas-launch-audit。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

saas-launch-audit 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

saas-launch-audit 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

saas-launch-audit 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

saas-launch-audit 用的人多吗?

公开数据是累计安装 815 次,近一周安装 35 次,下载 286 次,GitHub 11 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

saas-launch-audit 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
saas-launch-audit

saas launch 审核 · 本页

81511操作浏览器MIT有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

SaaS Launch Audit

A comprehensive, repeatable audit workflow for SaaS products approaching launch. Provides clear PASS/FAIL states with single actionable next steps.

Purpose

Validate launch readiness across 10 critical areas: validation, pricing, payments, AI economics, discovery, competitors, social presence, legal, trust infrastructure, and launch day operations.

Workflow Overview

  1. Discovery Survey - Gather project context (product type, audience, stage)
  2. Interactive Verification - Actively verify items using browser tools and web queries
  3. PASS/FAIL Report - Output each item with status and next step for failures
  4. Iterate - Re-run audit after fixes until all items pass

Discovery Survey

Before running the checklist, gather project context:

Questions to ask:

  1. Product type: B2B SaaS, B2C, Developer Tool, Marketplace, or Other?
  2. Target audience/ICP: Who is the primary customer?
  3. Current stage: Pre-MVP, Beta/Early Access, or Ready to Launch?
  4. Does the product use AI/ML that affects unit economics?
  5. Primary domain/URL to audit (if live)

Platform recommendations by product type:

  • Developer Tools: GitHub, Hacker News, Reddit, X, Dev.to
  • B2B SaaS: LinkedIn, Product Hunt, G2/Capterra
  • B2C/Consumer: X, TikTok, Instagram, YouTube
  • Indie Hacker: Product Hunt, Indie Hackers, X, Reddit

Audit Checklist (10 Sections)

1. Pre-Launch Validation

ItemVerification Method
50+ discovery calls completedAsk user
Pre-payment secured from 3+ customersAsk user
Problem validated (not just interest)Ask user for evidence
Manual service delivery testedAsk user

2. Pricing Strategy

ItemVerification Method
Pricing model selected (outcome/usage/hybrid)Ask user, verify NOT per-seat only
Value metric identifiedAsk user what outcome they charge for
3-4 tiers with clear differentiationBrowser: check pricing page
Annual contract incentives configuredBrowser: verify discount shown
Pricing page live and scannableBrowser: visual check

3. Payment Integration

ItemVerification Method
Stripe account activated (live keys)Ask user to confirm
Test transactions successfulAsk user
Webhook endpoints configuredAsk about endpoint URL
Billing portal accessibleBrowser: test billing portal link
Tax/KYC requirements metAsk user

4. AI Economics (if applicable)

ItemVerification Method
Inference costs calculated per actionAsk user for cost breakdown
AI costs < 30% of revenueCalculate: inference cost / revenue per user
Gross margin > 50% (AI-adjusted)Verify: traditional SaaS 80-90%, AI-first 50-65%
Margin protection strategy existsAsk about usage caps, tiering
Cost monitoring/alerts configuredAsk user

5. GEO & AI Discovery

ItemVerification Method
AI visibility audit completedWebSearch: query ChatGPT/Perplexity for brand
Structured data/FAQ schemaBrowser: check page source
Content uses authoritative languageBrowser: check for excessive hedging
SSR/pre-rendering for LLM crawlersAsk about rendering strategy
Server response time < 200msRun scripts/tech-audit.sh
Core Web Vitals passingCheck PageSpeed Insights
2026 Fact: AI agents account for ~33% of organic search activity. Only 12% of companies appear across ChatGPT, Perplexity, and Claude.

6. Competitor Intelligence

ItemVerification Method
3+ direct competitors identifiedAsk user to list
Competitor pricing documentedWebSearch: research each competitor
Differentiation statement writtenAsk user for positioning
AI visibility comparison doneWebSearch: query "[category] tools"
Citation sentiment analyzedReview AI responses for tone

7. Social Media & Launch Prep

ItemVerification Method
Primary platform profiles createdBrowser: check each platform
LinkedIn founder profile optimizedBrowser: review profile
X/Twitter presence establishedBrowser: check account
Product Hunt profile readyBrowser: check PH profile
400+ PH supporters lined upAsk user about DM outreach
Launch email sequence draftedAsk user
Content calendar for 30 daysAsk user
"Magic Triangle" readyAsk about founder content + outreach + retargeting

8. Legal & Compliance

ItemVerification Method
Privacy Policy liveBrowser: check /privacy or footer link
Terms of Service liveBrowser: check /terms or footer link
Cookie consent bannerBrowser: verify banner appears
DPA template ready (if B2B)Ask user

9. Trust Infrastructure

ItemVerification Method
Auth system production-readyAsk user about auth provider
Audit logging enabledAsk user
Error tracking configuredAsk about Sentry or similar
Uptime monitoring activeAsk user
Security headers configuredRun scripts/tech-audit.sh

10. Launch Day Readiness

ItemVerification Method
Support channels operationalBrowser: check support links
FAQ/Knowledge base publishedBrowser: check /help or /docs
Analytics tracking verifiedAsk about GA4, PostHog setup
Team roles assignedAsk user
Rollback plan documentedAsk user

Output Format

Report each item as:

[PASS] Item description

or

[FAIL] Item description
  → Next step: Single actionable instruction

Verification Tools

Browser Verification

Use mcp__claude-in-chrome__* tools to:

  • Navigate to pricing, terms, privacy pages
  • Check for cookie banners, structured data
  • Verify billing portal accessibility
  • Review social media profiles

Web Search Verification

Use WebSearch to:

  • Query AI assistants for brand visibility
  • Research competitor pricing
  • Check product mentions and sentiment

Script Verification

Use provided scripts:

  • scripts/tech-audit.sh - Core Web Vitals, SSL, security headers
  • scripts/competitor-visibility.sh - AI visibility comparison

Additional Resources

For detailed guidance, see the references directory:

  • references/pricing-2026.md - Outcome/usage-based pricing strategies, 2026 trends, tier design principles
  • references/payment-verification.md - Stripe go-live checklist
  • references/geo-visibility.md - GEO/AI discovery optimization, technical requirements, content strategies
  • references/launch-tactics.md - Product Hunt and social strategies, "Magic Triangle" approach
  • references/trust-infrastructure.md - Trust-as-a-Product requirements
  • references/metrics-2026.md - AI-first SaaS metrics (gross margins, cost optimization)
  • references/platform-guides.md - Platform-specific launch guides

Completion Criteria

The audit is complete when:

  1. All 10 sections have been evaluated
  2. Each item shows PASS or FAIL with next step
  3. User has a prioritized list of fixes

Re-run the audit after implementing fixes until all critical items pass.

Key 2026 Insights

  1. Pricing: Per-seat is declining (126% YoY growth in credit models). Use outcome-based or hybrid models.
  2. AI Economics: Target AI costs < 30% of revenue. Traditional SaaS: 80-90% margins. AI-first: 50-65% margins.
  3. Discovery: AI agents now ~33% of organic search. Only 12% of companies appear across all AI platforms.
  4. Trust: Trust-as-a-Product is a differentiator, not just compliance.
  5. Product Hunt: 26.5% of launches get 500-1000 users. 49% fail by relying on PH alone.

平台分布

Claude Code

26.71%
按下载量换算76

windsurf

25.69%
按下载量换算73

OpenCode

17.36%
按下载量换算50

Antigravity

13.98%
按下载量换算40

Gemini CLI

7.41%
按下载量换算21

Cursor

3.27%
按下载量换算9

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。

继续浏览同类 Skills