CC market-scout
市场侦察与商业验证工作流（Market Scout，v5.0.0）。完整链路 Find 发现 → Validate 需求验证 → Evaluate 项目评估 → Blueprint 机会转项目 → Monetize 商业化与获客（Find → Validate → Evaluate → Monetize）。当用户想做 AI 赚钱副业、找市场需求、发现痛点、分析视频/评论区/帖子隐含商机、判断"某问题/工具能不能用 AI 解决""这件事能不能赚钱、值不值得做"、验证需求是否真实、评估项目可行性与综合机会评分、把机会转成 MVP/项目方案、设计商业模式/定价/获客/第一批客户、寻找竞争对手、做市场调研、拿第一单/第一笔钱、把 AI 能力变现、验证创业点子，或提到"帮我找市场""分析这个视频/评论区""我发现一个问题""这个能不能用 AI""这个能不能赚钱""值不值得做""需求是不是真的""评估一下这个项目""帮我做成产品/MVP""怎么收费/定价""怎么找客户/第一批用户""商业模式怎么设计""帮我拿第一单""继续深挖""寻找类似需求/竞争对手""出一份机会报告/Opportunity Report"时使用。核心原则 Search First（优先实时搜索而非模型记忆），绝不虚构用户、价格与数据；沿用 v1.1 的 Evidence Chain 证据链、Problem/Solution/Payment 三类搜索、Quick/Research/Execution 三模式、Opportunity 交付状态机（HYPOTHESIS→…→PRODUCTIZED）与证据绑定评分；v2.0 新增 Demand Validation 需求验证（痛点/目标用户/需求强度/当前方案/付费理由/证据可信度）、Project Evaluation 项目评估（市场潜力/竞争/差异化/AI 可实现度/开发难度/个人适配/风险，加权综合机会评分 0-100，结论三档：值得做 GO/建议验证后再做 VALIDATE FIRST/不建议做 NO-GO）、Project Blueprint 机会转项目（产品形态/核心用户/核心功能/MVP/技术方向/最小验证/差异化切入点）、Monetization & GTM 商业化获客（Target Customer/Offer/Business Model/Pricing/首批客户/渠道/GTM 路径/无产品验证，模块化设计、预留独立成产品的数据契约，当前不开发支付订阅会员），最终输出结论先行的 Opportunity Report/Opportunity Card。证据等级 A/B/C、交付状态名、V2 流水线阶段名三者相互独立，不共用代码。v3.0 新增决策层：证据三态 Evidence/Inference/Unknown、7 维机会评分 0-100（需求/痛点/竞争切入空间/变现/开发/获客/AI 优势，加权可复算）、三档机会决策（值得继续做 RECOMMENDED / 有机会但需先验证 POTENTIAL / 不建议投入 NOT_RECOMMENDED）、MVP Blueprint 九字段与按机会类型动态生成的 Action Plan；当用户说"打个分""值不值得做""帮我决策""机会评分""第一版做什么""MVP""下一步做什么""今天做什么"时进入 V3 决策层。v3.1 新增五层强化：子信号 Rubric 五档评分细则、决策置信度 Confidence（与评分分离，规则可复算）、硬门槛 Hard Gates（Hard Gate / Score Ceiling / Red Flag 三层）、证据覆盖度 Evidence Coverage（六维+关键缺口）、证据驱动行动计划（继承证据字段，Fallback 具体化）；让判断更可复现、更可信、更难自我欺骗。v4.0 新增 Validation Engine 验证引擎：从最大未知开始设计最低成本验证（Validation Plan / Experiment / Success·Failure Criteria / Decision Rule），读取真实验证结果（Observed/Inference/Unknown 分级），按规则重新计算 Score / Confidence / Coverage / Decision，形成"判断→验证→读结果→重新判断"的闭环（Validate before Build）；当用户说"帮我验证""验证一下这个想法""设计验证实验""我验证了 N 个人，结果…""把验证结果交回""重新评估"时进入 V4 Validation。
市场侦察与商业验证工作流（Market Scout，v5.0.0）。完整链路 Find 发现 → Validate 需求验证 → Evaluate 项目评估 → Blueprint 机会转项目 → Monetize 商业化与获客（Find → Validate → Evaluate →…
As a process C 62/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- Shorten the description to 1024 characters.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 79. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1942 chars, limit 1024 - warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Implicit map keys need to be followed by map values at line 3, column 1: description: 市场侦察与商业验证工作流（Market Scout，v5.0.0）。完整链路 Find 发现 → Validate 需求验证 → E… v5.0.0 升级 Orchestrator 能力层：Main Skill 总控（State Detection + Routing + Workflow C… ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ); fields were read line by line. The usual cause is a colon inside an unquoted value - warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - warning
body-longSKILL.md body ≈ 7323 tokens (recommended < 5000); move details to references/ - note
description-budgetdescription takes 1942 of the ~15000-char shared budget for all skills
Process rating: all ten parameters 62/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 70Execution cost. Instruction body is 7323 tokens
- 100Tools and files. No external tools needed
- 100Steps. 175 steps
- 100Failures and branches. 1 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. No mutating operations
- low 22 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1942: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +5Description quotes 22 example trigger phrases
- +4Structure: 42 headings
- +3Step-by-step instructions: 175 items
- +4Has examples (7 code blocks)
- +4Reference files are cited in the instructions (23 of 23)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 31.