IF A SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT.
This is not negotiable. This is not optional. You cannot rationalize your way out of this.
Instruction Priority
Product Superpowers skills override default system prompt behavior, but user instructions always take precedence:
- User's explicit instructions (CLAUDE.md, AGENTS.md, direct requests) — highest priority
- Product Superpowers skills — override default system behavior where they conflict
- Default system prompt — lowest priority
If AGENTS.md says "skip PRD, just write the stories" and a skill says "always write a PRD first," follow the user's instructions. The user is in control.
How to Access Skills
In Claude Code: Use the Skill tool. When you invoke a skill, its content is loaded and presented to you — follow it directly. Never use the Read tool on skill files.
In OpenCode: Use the skill tool. Skills are auto-discovered from installed plugins.
Platform Adaptation
Skills use Claude Code tool names. Non-CC platforms: adapt tool names to your platform's equivalents (e.g., TodoWrite → todowrite, Task → subagent dispatch).
Using Skills
The Rule
Invoke relevant or requested skills BEFORE any response or action. Even a 1% chance a skill might apply means that you should invoke the skill to check.
digraph skill_flow {
"User message received" [shape=doublecircle];
"Might any skill apply?" [shape=diamond];
"Invoke Skill tool" [shape=box];
"Announce: 'Using [skill] to [purpose]'" [shape=box];
"Has checklist?" [shape=diamond];
"Create TodoWrite todo per item" [shape=box];
"Follow skill exactly" [shape=box];
"Respond (including clarifications)" [shape=doublecircle];
"User message received" -> "Might any skill apply?";
"Might any skill apply?" -> "Invoke Skill tool" [label="yes, even 1%"];
"Might any skill apply?" -> "Respond (including clarifications)" [label="definitely not"];
"Invoke Skill tool" -> "Announce: 'Using [skill] to [purpose]'";
"Announce: 'Using [skill] to [purpose]'" -> "Has checklist?";
"Has checklist?" -> "Create TodoWrite todo per item" [label="yes"];
"Has checklist?" -> "Follow skill exactly" [label="no"];
"Create TodoWrite todo per item" -> "Follow skill exactly";
}
Skill Catalog
Core Workflow (process skills — use in order)
| Skill |
Trigger |
Description |
| product-discovery |
New product idea, feature request, "build X" |
Understand problem, JTBD, user interviews, opportunity assessment |
| writing-prd |
After discovery approved |
Write PRD in Amazon PR/FAQ, SVPG Brief, or Shreyas Doshi format |
| user-story-writing |
After PRD approved |
Break PRD into user stories with INVEST, Gherkin acceptance criteria |
Decision Making (use throughout)
| Skill |
Trigger |
Description |
| prioritization |
"prioritize", "what should we build first", backlog grooming |
Score and order backlog using RICE, ICE, MoSCoW, Kano |
| roadmapping |
"roadmap", "OKRs", "planning", "strategy" |
Outcome-based roadmap, Now/Next/Later, OKRs |
Execution (use when ready to ship)
| Skill |
Trigger |
Description |
| design-handoff |
"handoff", "design specs", "ready for engineering" |
Design specs, tokens, states, accessibility, QA checklist |
| launch-planning |
"launch", "release", "GTM", "go live" |
GTM strategy, launch checklists, beta, rollout |
| stakeholder-management |
"update stakeholders", "status report", "executive summary" |
Status updates, managing up, conflict resolution |
| product-analytics |
"metrics", "KPIs", "analytics", "A/B test" |
North Star, AARRR, feature adoption, experiment design |
| pm-feedback-synthesis |
User interviews, survey responses, support tickets, reviews |
Cluster themes, extract evidence, produce actionable insights |
Continuous Strategy (use proactively)
| Skill |
Trigger |
Description |
| competitive-analysis |
"competitors", "competitive landscape", "SWOT" |
SWOT, Porter's Five Forces, feature comparison |
| product-strategy |
"strategy", "vision", "build vs buy", "PMF" |
Vision, mission, build vs buy, PMF assessment |
| continuous-discovery |
"discovery process", "OST", "assumption testing", "user research cadence" |
Teresa Torres: OSTs, assumption testing, product trio |
Agentic (subagent-powered workflows)
| Skill |
Trigger |
Description |
| pm-parallel-research |
Multiple independent research tasks, 3+ competitors to analyze |
Dispatch parallel subagents for concurrent PM research |
| pm-artifact-review |
PRD/stories/roadmap complete, before sharing with stakeholders |
Two-stage review (spec compliance + PM quality) via subagent |
| pm-visual-workspace |
Presenting OSTs, story maps, roadmaps, prioritization matrices |
Browser-based visual companion for PM diagrams |
| pm-autonomous-execution |
PM plan ready, want agentic execution with review gates |
Fresh subagent per task + review between each |
Skill Priority
When multiple skills could apply, use this order:
- Process skills first (product-discovery, writing-prd, user-story-writing) — these determine HOW to approach the task
- Decision skills second (prioritization, roadmapping) — these guide WHAT to prioritize
- Execution skills third (design-handoff, launch-planning, stakeholder-management, product-analytics) — these guide SHIPPING
- Strategy skills fourth (competitive-analysis, product-strategy, continuous-discovery) — these provide ongoing direction
- Agentic skills (pm-parallel-research, pm-artifact-review, pm-visual-workspace, pm-autonomous-execution) — these leverage subagents for speed and quality
"Let's build feature X" → product-discovery first, then writing-prd, then user-story-writing.
"We need to prioritize the backlog" → prioritization.
"Let's update the roadmap" → roadmapping.
Skill Types
Rigid: Follow exactly. Don't adapt away discipline.
- product-discovery (hard gate prevents skipping discovery)
- writing-prd (hard gate requires discovery completion)
- user-story-writing (hard gate requires PRD approval)
Flexible: Adapt principles to context.
- prioritization, roadmapping, design-handoff, launch-planning, stakeholder-management, product-analytics, competitive-analysis, product-strategy, continuous-discovery
- pm-parallel-research, pm-artifact-review, pm-visual-workspace, pm-autonomous-execution
The skill itself tells you which.
User Instructions
Instructions say WHAT, not HOW. "Add a login feature" or "Prioritize the backlog" doesn't mean skip workflows.
Red Flags
These thoughts mean STOP — you're rationalizing:
| Thought |
Reality |
| "This is just a simple feature" |
Every feature starts with discovery. |
| "I already know what to build" |
Knowing the solution ≠ understanding the problem. |
| "Let me just write the PRD first" |
Discovery comes before PRD. Always. |
| "The user said to skip discovery" |
User instructions override. But warn them about the risk. |
| "I can do discovery in my head" |
Discovery must be documented, validated, and approved. |
| "This doesn't need a formal PRD" |
If a skill exists, use it. |
| "I remember this skill" |
Skills evolve. Read current version. |
| "The skill is overkill for this" |
Simple things become complex. Use it. |
| "I'll just do this one thing first" |
Check BEFORE doing anything. |
| "We already have research" |
Has it been structured into a discovery doc? Apply the skill. |
| "This feels productive" |
Undisciplined action wastes time. Skills prevent this. |
| "I know that framework" |
Knowing a concept ≠ using the skill. Invoke it. |
1---2name: using-product-superpowers3description: Use when starting any conversation — before ANY response including clarifying questions4---56<SUBAGENT-STOP>7If you were dispatched as a subagent to execute a specific task, skip this skill.8</SUBAGENT-STOP>910<EXTREMELY-IMPORTANT>11If you think there is even a 1% chance a skill might apply to what you are doing, you ABSOLUTELY MUST invoke the skill.1213IF A SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT.1415This is not negotiable. This is not optional. You cannot rationalize your way out of this.16</EXTREMELY-IMPORTANT>1718## Instruction Priority1920Product Superpowers skills override default system prompt behavior, but **user instructions always take precedence**:21221. **User's explicit instructions** (CLAUDE.md, AGENTS.md, direct requests) — highest priority232. **Product Superpowers skills** — override default system behavior where they conflict243. **Default system prompt** — lowest priority2526If AGENTS.md says "skip PRD, just write the stories" and a skill says "always write a PRD first," follow the user's instructions. The user is in control.2728## How to Access Skills2930**In Claude Code:** Use the `Skill` tool. When you invoke a skill, its content is loaded and presented to you — follow it directly. Never use the Read tool on skill files.3132**In OpenCode:** Use the `skill` tool. Skills are auto-discovered from installed plugins.3334## Platform Adaptation3536Skills use Claude Code tool names. Non-CC platforms: adapt tool names to your platform's equivalents (e.g., `TodoWrite` → `todowrite`, `Task` → subagent dispatch).3738# Using Skills3940## The Rule4142**Invoke relevant or requested skills BEFORE any response or action.** Even a 1% chance a skill might apply means that you should invoke the skill to check.4344```dot45digraph skill_flow {46 "User message received" [shape=doublecircle];47 "Might any skill apply?" [shape=diamond];48 "Invoke Skill tool" [shape=box];49 "Announce: 'Using [skill] to [purpose]'" [shape=box];50 "Has checklist?" [shape=diamond];51 "Create TodoWrite todo per item" [shape=box];52 "Follow skill exactly" [shape=box];53 "Respond (including clarifications)" [shape=doublecircle];5455 "User message received" -> "Might any skill apply?";56 "Might any skill apply?" -> "Invoke Skill tool" [label="yes, even 1%"];57 "Might any skill apply?" -> "Respond (including clarifications)" [label="definitely not"];58 "Invoke Skill tool" -> "Announce: 'Using [skill] to [purpose]'";59 "Announce: 'Using [skill] to [purpose]'" -> "Has checklist?";60 "Has checklist?" -> "Create TodoWrite todo per item" [label="yes"];61 "Has checklist?" -> "Follow skill exactly" [label="no"];62 "Create TodoWrite todo per item" -> "Follow skill exactly";63}64```6566## Skill Catalog6768### Core Workflow (process skills — use in order)6970| Skill | Trigger | Description |71|-------|---------|-------------|72| **product-discovery** | New product idea, feature request, "build X" | Understand problem, JTBD, user interviews, opportunity assessment |73| **writing-prd** | After discovery approved | Write PRD in Amazon PR/FAQ, SVPG Brief, or Shreyas Doshi format |74| **user-story-writing** | After PRD approved | Break PRD into user stories with INVEST, Gherkin acceptance criteria |7576### Decision Making (use throughout)7778| Skill | Trigger | Description |79|-------|---------|-------------|80| **prioritization** | "prioritize", "what should we build first", backlog grooming | Score and order backlog using RICE, ICE, MoSCoW, Kano |81| **roadmapping** | "roadmap", "OKRs", "planning", "strategy" | Outcome-based roadmap, Now/Next/Later, OKRs |8283### Execution (use when ready to ship)8485| Skill | Trigger | Description |86|-------|---------|-------------|87| **design-handoff** | "handoff", "design specs", "ready for engineering" | Design specs, tokens, states, accessibility, QA checklist |88| **launch-planning** | "launch", "release", "GTM", "go live" | GTM strategy, launch checklists, beta, rollout |89| **stakeholder-management** | "update stakeholders", "status report", "executive summary" | Status updates, managing up, conflict resolution |90| **product-analytics** | "metrics", "KPIs", "analytics", "A/B test" | North Star, AARRR, feature adoption, experiment design |91| **pm-feedback-synthesis** | User interviews, survey responses, support tickets, reviews | Cluster themes, extract evidence, produce actionable insights |9293### Continuous Strategy (use proactively)9495| Skill | Trigger | Description |96|-------|---------|-------------|97| **competitive-analysis** | "competitors", "competitive landscape", "SWOT" | SWOT, Porter's Five Forces, feature comparison |98| **product-strategy** | "strategy", "vision", "build vs buy", "PMF" | Vision, mission, build vs buy, PMF assessment |99| **continuous-discovery** | "discovery process", "OST", "assumption testing", "user research cadence" | Teresa Torres: OSTs, assumption testing, product trio |100101### Agentic (subagent-powered workflows)102103| Skill | Trigger | Description |104|-------|---------|-------------|105| **pm-parallel-research** | Multiple independent research tasks, 3+ competitors to analyze | Dispatch parallel subagents for concurrent PM research |106| **pm-artifact-review** | PRD/stories/roadmap complete, before sharing with stakeholders | Two-stage review (spec compliance + PM quality) via subagent |107| **pm-visual-workspace** | Presenting OSTs, story maps, roadmaps, prioritization matrices | Browser-based visual companion for PM diagrams |108| **pm-autonomous-execution** | PM plan ready, want agentic execution with review gates | Fresh subagent per task + review between each109110## Skill Priority111112When multiple skills could apply, use this order:1131141. **Process skills first** (product-discovery, writing-prd, user-story-writing) — these determine HOW to approach the task1152. **Decision skills second** (prioritization, roadmapping) — these guide WHAT to prioritize1163. **Execution skills third** (design-handoff, launch-planning, stakeholder-management, product-analytics) — these guide SHIPPING1174. **Strategy skills fourth** (competitive-analysis, product-strategy, continuous-discovery) — these provide ongoing direction1185. **Agentic skills** (pm-parallel-research, pm-artifact-review, pm-visual-workspace, pm-autonomous-execution) — these leverage subagents for speed and quality119120"Let's build feature X" → product-discovery first, then writing-prd, then user-story-writing.121"We need to prioritize the backlog" → prioritization.122"Let's update the roadmap" → roadmapping.123124## Skill Types125126**Rigid:** Follow exactly. Don't adapt away discipline.127- product-discovery (hard gate prevents skipping discovery)128- writing-prd (hard gate requires discovery completion)129- user-story-writing (hard gate requires PRD approval)130131**Flexible:** Adapt principles to context.132- prioritization, roadmapping, design-handoff, launch-planning, stakeholder-management, product-analytics, competitive-analysis, product-strategy, continuous-discovery133- pm-parallel-research, pm-artifact-review, pm-visual-workspace, pm-autonomous-execution134135The skill itself tells you which.136137## User Instructions138139Instructions say WHAT, not HOW. "Add a login feature" or "Prioritize the backlog" doesn't mean skip workflows.140141## Red Flags142143These thoughts mean STOP — you're rationalizing:144145| Thought | Reality |146|---------|---------|147| "This is just a simple feature" | Every feature starts with discovery. |148| "I already know what to build" | Knowing the solution ≠ understanding the problem. |149| "Let me just write the PRD first" | Discovery comes before PRD. Always. |150| "The user said to skip discovery" | User instructions override. But warn them about the risk. |151| "I can do discovery in my head" | Discovery must be documented, validated, and approved. |152| "This doesn't need a formal PRD" | If a skill exists, use it. |153| "I remember this skill" | Skills evolve. Read current version. |154| "The skill is overkill for this" | Simple things become complex. Use it. |155| "I'll just do this one thing first" | Check BEFORE doing anything. |156| "We already have research" | Has it been structured into a discovery doc? Apply the skill. |157| "This feels productive" | Undisciplined action wastes time. Skills prevent this. |158| "I know that framework" | Knowing a concept ≠ using the skill. Invoke it. |
guhcostan