pm-skill-creator

Files (1)

• SKILL.md

SKILL.md (first lines)

## Purpose

Walk through the full skill design process interactively — from raw idea or content to a structured, repo-compliant SKILL.md draft. Asks adaptive questions to determine skill type, scope, structure, and content, then generates a ready-to-validate draft.

This skill sits with you and figures out *what* to build before you build it — shaping a raw idea or content into the right structure through conversation, then generating a draft you can validate.

## Key Concepts

### The Three Skill Types (Decision Criteria)

- **Component**: One artifact or template. Self-contained. Gets referenced by other skills. Ask: "Is this a thing someone creates?"
- **Interactive**: Guided conversation with adaptive questions and enumerated recommendations. Ask: "Does this require back-and-forth to be useful?"
- **Workflow**: Multi-phase orchestration referencing other skills. Ask: "Does this coordinate multiple activities across steps?"

### Skill Anatomy (Non-Negotiable Sections)

Every skill requires these sections in order:
1. **Purpose** — What it does + when to use it (outcome-focused)
2. **Key Concepts** — Frameworks, definitions, mental models
3. **Application** — Step-by-step instructions an agent can follow
4. **Examples** — At least one concrete, specific example
5. **Common Pitfalls** — Named failure modes with consequences and fixes
6. **References** — Related skills, external frameworks, source material

### Metadata Constraints

- `name`: lowercase kebab-case, ≤ 64 characters
- `description`: ≤ 200 characters, must include trigger cue ("Use when...")
- `intent`: longer repo-facing summary (no character limit)
- `type`: one of `component`, `interactive`, `workflow`

### Facilitation Protocol

Run this skill as a guided conversation using this interaction protocol:
- Open with a one-line heads-up, then offer an entry mode: **Guided** (one question at a time), **Context dump** (the user pastes everything up front), or **Best guess** (proceed from the topic alone).
- Ask **one question per turn** in plain language; wait for the answer before asking the next. Never bundle multiple questions.
- Label progress on each question (e.g., `Context Q1/5`).
- With each question, offer **numbered response options and a recommended default**; include `Other (specify)` when useful.
- Handle interruptions gracefully and allow the user to pause and resume.
- At decision points, present **numbered recommendations** rather than a single take.

## Application

This interactive skill asks **up to 5 adaptive questions**, then delivers a **complete SKILL.md draft** with frontmatter, all required sections, and repo-compliant structure.

---

### Step 0: Session Start

**Agent says:**

"I'll help you design a new PM skill from scratch. This takes about 10-15 minutes and up to 5 questions. How do you want to start?

1. **Guided** — I'll ask questions one at a time and build the skill from your answers (recommended)
2. **Context dump** — Paste your raw content, notes, or framework and I'll propose a skill structure
3. **Best guess** — Tell me just the topic and I'll draft a skill you can refine"

---

### Question 1: What's the Raw Material? (Q1/5)

**Agent asks:**

"What are we turning into a skill? Give me whatever you have."

1. **A framework or mental model** — A structured way of thinking about a PM problem (e.g., prioritization matrix, decision tree)
2. **A template or artifact** — A deliverable PMs create (e.g., PRD section, positioning statement, epic format)
3. **A process or workflow** — A multi-step method for completing a PM task (e.g., discovery sprint, roadmap planning)
4. **A coaching or advisory topic** — A domain where PMs need guided, adaptive help (e.g., stakeholder navigation, pricing decisions)

**Or describe what you have in your own words.**

**Agent note:** If the user pastes raw content instead of choosing an option, analyze the content and infer the answer. Confirm your interpretation before proceeding.

---

### Question 2: Skill Type Decision (Q2/5)

**Based on Q1 answer, agent recommends a type and confirms:**

*If Q1 = Framework or Template:*

"This sounds like a **component skill** — a self-contained artifact or reference. It would include a template, quality criteria, and examples. Does that fit, or is there a conversational/adaptive element I'm missing?

1. **Yes, component** — It's a standalone deliverable or reference
2. **Actually, it needs conversation** — Users need guided questions to use it well (→ interactive)
3. **It's bigger than one artifact** — It orchestrates multiple steps or other skills (→ workflow)"

*If Q1 = Process or Workflow:*

"This sounds like a **workflow skill** — a multi-phase process. Does it reference or orchestrate other discrete skills/artifacts, or is it more of a guided conversation?

1. **Yes, workflow** — It has distinct phases and may reference other skills
2. **It's more conversational** — The value comes from adaptive questions and recommendations (→ interactive)
3. **It's simpler than I described** — Really it's one artifact with steps (→ component)"

*If Q1 = Coaching or Advisory:*

"This sounds like an **interactive skill** — guided questions that adapt based on answers, ending with enumerated recommendations. Sound right?

1. **Yes, interactive** — It needs back-and-forth to be useful
2. **It's more of a reference** — Users just need the framework, not a conversation (→ component)
3. **It's a full process** — Multiple phases, orchestrates other skills (→ workflow)"

---

### Question 3: Scope and Naming (Q3/5)

**Agent asks:**

"What should we call this skill? I need two things:

**a) A working name** — lowercase-kebab-case, max 64 characters (e.g., `feature-investment-advisor`, `user-story`, `discovery-process`)

**b) A one-sentence description** — What it does + when to use it. Must fit in 200 characters and start with a verb or 'Use when...'

Give me your best attempt and I'll tighten it if needed. Or just describe the skill's purpose and I'll propose both."

**Agent note:** Validate the name format (kebab-case, ≤ 64 chars) and description length (≤ 200 chars) before proceeding. If either fails, suggest a fix.

From Infrasity-Labs/dev-gtm-claude-skills, MIT.

More from Infrasity-Labs/dev-gtm-claude-skills