技能备份 - 2026-04-15 (40个技能)

skills/product-manager-skills/examples/director-coaching.md

@@ -0,0 +1,33 @@
+# Example: Director Readiness Coaching
+
+## Scenario
+
+You are moving from senior IC or group PM work into Director-level expectations.
+
+## Prompt
+
+```text
+I'm a senior PM managing two PMs, strong on execution, weak on org influence, and interviewing for Director roles in 3 months. Coach me.
+```
+
+## Example Output Excerpt
+
+```text
+Diagnosis
+- You are operating strongly at team altitude and inconsistently at org altitude.
+- Your likely interview risk is not execution depth. It is proving that you can shape priorities across teams and influence without direct authority.
+
+Gaps to close in the next 90 days
+- Build 3 stories that show portfolio tradeoffs, not just feature delivery.
+- Practice framing decisions in terms of company bets, not just team output.
+- Create a weekly visibility loop with engineering, design, and GTM stakeholders.
+
+Recommended plan
+- Weeks 1-2: inventory past projects and rewrite 5 stories using org-level impact language.
+- Weeks 3-6: run mock interviews focused on conflict, strategy, and delegation.
+- Weeks 7-12: ship one visible cross-functional improvement and capture evidence of influence.
+```
+
+## Why This Gets Reused
+
+Career coaching becomes sticky when the advice is specific to role altitude, not generic motivation.

skills/product-manager-skills/examples/prd-review.md

@@ -0,0 +1,36 @@
+# Example: PRD Review And Pushback
+
+## Scenario
+
+You already have a PRD draft, but you want a strong PM peer review instead of grammar cleanup.
+
+## Prompt
+
+```text
+Review this PRD for a notification preferences center. Flag solution smuggling, weak metrics, overscoping, and delivery risk. Be opinionated.
+```
+
+## Example Output Excerpt
+
+```text
+Findings
+- The problem statement is solution-smuggled. "Users need a preferences dashboard" starts from UI, not from the user problem.
+- Success metrics are incomplete. There is no baseline, no target, and no guardrail metric.
+- Scope is too broad for one release: channel controls, digest preferences, quiet hours, admin policy, and migration are multiple slices.
+- User stories are horizontal. API work, UI work, and notification service updates are separated instead of being cut by user value.
+
+Suggested reframing
+- Problem: users receive too many irrelevant notifications and lack confidence they can control frequency without missing critical updates.
+- First release: account-level email preferences + unsubscribe rescue flow + instrumentation for preference changes and unsubscribe reduction.
+
+Recommended next step
+- Rewrite the problem statement and success metrics first. Do not expand solution detail until scope is sliced vertically.
+```
+
+## What Good Looks Like
+
+The review should not stop at criticism. It should:
+
+- name the anti-pattern directly
+- explain the consequence
+- propose a thinner, more testable slice

skills/product-manager-skills/examples/saas-health-diagnostic.md

@@ -0,0 +1,37 @@
+# Example: SaaS Health Diagnostic
+
+## Scenario
+
+You want a fast diagnosis from a small set of numbers, not a finance lecture.
+
+## Prompt
+
+```text
+Analyze these metrics: MRR $50k, 500 customers, gross margin 80%, monthly churn 8%, CAC $500.
+```
+
+## Example Output Excerpt
+
+```text
+Diagnosis
+- 8% monthly churn compounds to roughly 63% annual churn. This is a red flag, not a mild retention issue.
+- ARPA is about $100/month. With 80% gross margin and 8% monthly churn, better LTV is about $1,000.
+- LTV:CAC is about 2:1. That is below a healthy SaaS target, even though payback is workable at about 6.25 months.
+- The bottleneck is retention, not acquisition efficiency. Scaling paid acquisition before fixing churn will hide the real problem.
+
+Likely root causes to inspect next
+- Weak activation in the first 14 days
+- Low repeat value for the core job
+- Poor fit in one or more acquisition cohorts
+
+Recommended next step
+- Pull 3 monthly cohorts, compare retention at day 30 / 60 / 90, and isolate the biggest drop-off before changing spend.
+```
+
+## Why This Matters
+
+The useful part is not just the math. The skill converts the math into a decision:
+
+- what is actually broken
+- what not to optimize yet
+- what analysis should come next