Skill AI למנהלי מוצר (PM) ל-Claude

📖 מה ה-Skill הזה כולל

מתי להשתמש

"AI for PM", "Product manager AI tools", "AI in product development".

הוראות עבודה

1. PM's AI Use Cases

Daily

PRD drafts — Claude with template.
User interview summaries — Otter + Claude.
Customer feedback synthesis — Claude on reviews.
Slack / email drafts.

Weekly

Sprint planning assist.
Roadmap docs — AI structures.
Stakeholder updates — auto-generate.

Monthly+

Competitive analysis — AI scrape + analyze.
Quarterly planning — AI synthesizes inputs.

2. Stack — PM Personal

Claude Pro ($20/m) — main thinking partner.
Perplexity Pro ($20) — research.
Otter.ai ($16-30) — meeting transcripts.
Gamma ($10) — instant slides.
v0 ($20) — UI mocks.

3. PRD Workflow

1. Idea + initial brainstorm (Claude).
2. Research competitors (Perplexity).
3. User interviews summarized (Otter + Claude).
4. PRD draft (Claude with template).
5. Iterate with team feedback.
6. Final PRD.

Time: Days → Hours.

4. PRD Template Prompt

You are a Senior Product Manager.

Generate a PRD for the following feature:
[DESCRIPTION]

Include:
1. Problem statement
2. User personas
3. User stories (5-8)
4. Acceptance criteria
5. Success metrics
6. Edge cases / Risks
7. Timeline estimate
8. Dependencies

Format: Markdown.
Tone: Concise, decision-focused.

5. User Research Synthesis

Workflow

10 user interviews → transcribed.
Upload to Claude.
Prompt: "Synthesize themes, common pain points, surprises, top quotes."
Output: research report in 30 min.

6. Roadmap Planning

AI Helps

Prioritization (RICE scoring with AI assist).
Dependency mapping.
Effort estimation.
Roadmap visualization.

7. Building AI Features in Product

Decision Framework

Customer-facing: Higher bar, careful UX.
Internal tools: Easier to ship.
AI-native or AI-enhanced: Different strategies.

Common AI Features 2026

Smart search (RAG).
Auto-summarization.
Chatbots.
Content generation.
Recommendations.
Prediction (churn, conversion).

8. PM-Engineering Collaboration on AI

PM's Role

Define problem.
Set guardrails.
Quality benchmarks.
User research.

Engineer's Role

Choose architecture.
Implementation.
Performance.
Reliability.

Together

Prompt engineering.
Eval design.
Fail mode planning.

9. AI Feature Discovery

Process

Identify repetitive user task.
Hypothesize: "Can AI do this 80% well?"
Spike POC (2-week sprint).
User test.
If 80%+ success → ship.
If not → iterate or kill.

10. Common AI Product Mistakes

❌ AI for the sake of AI — no real value. ❌ Over-promising — "AI does everything!" ❌ No fallback — AI fails, feature broken. ❌ Hidden AI — users discover, lose trust. ❌ No measurement — AI feature value unknown.

11. PRD for AI Feature — Special Sections

Model selection + reasoning.
Fallback when AI unavailable.
Confidence thresholds.
Human-in-loop triggers.
Privacy considerations.
Cost projections.
Quality metrics.

12. Israeli PM Specifics

Hebrew product — AI quality varies.
Israeli market data — limited.
Tech-savvy users — early AI adopters.

13. אסיים בהמלצה.

פרומפט לדוגמה

PM of B2B SaaS HR Tech. AI tools to adopt.

PRD for AI summarization feature. Template.

Decide: build AI feature or wait?

📥 התקנה בחצי דקה

1. הורד ופתח את קובץ ה-ZIP — תקבל תיקייה בשם ai-for-pm.

2. ב-Claude Code: העבר את התיקייה אל ~/.claude/skills/.
באפליקציה (Claude / Cowork): הגדרות ← Capabilities ← Skills ← העלאה.

3. בקש מ-Claude את מה שצריך בעברית — הוא יפעיל את ה-skill לבד כשזה רלוונטי.