agentleFS
drvossdrvoss/everything-copilot-cli/skills/product/create-prd/SKILL.md

create-prd

Skill48 starsChanged 17 days agoMITView on GitHub

Use when you're ready to define a feature before building it — generates a structured PRD using Jobs-to-be-Done thinking, success metrics, and explicit scope boundaries.

44 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use prioritize-features, brainstorm-okrs and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • skytiger6724Skill

    prioritize-features

    Prioritize a backlog of feature ideas based on impact, effort, risk, and strategic alignment with top 5 recommendations. Use when prioritizing a feature backlog, making scope decisions, or ranking product ideas.

    Also in 4 projects that use this

  • skytiger6724Skill

    brainstorm-okrs

    Brainstorm team-level OKRs aligned with company objectives — qualitative objectives with measurable key results. Use when setting quarterly OKRs, aligning team goals with company strategy, drafting objectives, or learning how to write effective OKRs.

    Also in 4 projects that use this

  • miovragSkill

    strategy-red-team

    Red-team a PRD, roadmap, or strategy by attacking its load-bearing assumptions before reality does. Steelmans then attacks each claim, ranks failure modes by impact × likelihood × cheapness-to-test, and returns the cheapest test and kill criteria for each. Use when stress-testing a plan, pressure-testing a strategy, challenging assumptions, or preparing a doc for executive review.

    Also in 3 projects that use this

  • launchapp-devSkill

    interview-script

    Create a structured user interview script with warm-up, core exploration, and wrap-up sections. Use when preparing for user research interviews to ensure consistent, insightful conversations.

    Also in 4 projects that use this

  • About-IntelligenceSkill

    beachhead-segment

    Identify the first beachhead market segment for a product launch.

    Also in 4 projects that use this

  • nandanosqlSkill

    customer-journey-map

    Map the complete customer experience across all touchpoints using the NNGroup framework. Identifies pain points, emotions, and opportunities at every stage of the user journey.

    Also in 5 projects that use this

  • +5 more from skytiger6724

The file

We split this skill into the rules it gives an agent

11 rules. 0 are common, 10 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 42 edited versions we read keep these.

  • 10

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before starting a new feature — to align engineering, product, and stakeholders on scope

    line 14
  2. Distinctive:

    When a GitHub Issue or Slack request needs formal requirements before implementation

    line 15
  3. Distinctive:

    When writing a PRD from scratch for a feature you're about to build solo

    line 16
  4. Distinctive:

    When the current conversation already contains enough context to synthesize a first PRD draft

    line 18
  5. Distinctive:

    Before formalizing requirements, briefly verify that the feature is worth building.

    line 25
  6. Very specific:

    If any answer is "unknown", record it in the PRD's Open Questions section and tag it needs-validation-before-implementation rather than …

    line 53
  7. Distinctive:

    If the current conversation, issue, or spec thread already contains enough information, draft from that context first.

    line 65
  8. Distinctive:

    Before writing requirements, anchor in the user's actual goal:

    line 93
  9. Distinctive:

    If your workflow uses GitHub Issues as the system of record, link the PRD from the issue or publish the summary there so implementation and …

    line 183
  10. Distinctive:

    Synthesize before interviewing again: if the conversation already answers the key questions, produce the first draft and ask only targeted …

    line 230
  11. Distinctive:

    PRD ≠ design spec: leave implementation details out.

    line 234

“Common” is measured: kept by at least 6 in 10 of the 42 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Create PRD

When to Use

  • Before starting a new feature — to align engineering, product, and stakeholders on scope
  • When a GitHub Issue or Slack request needs formal requirements before implementation
  • When writing a PRD from scratch for a feature you're about to build solo
  • As a handoff document: from discovery → design → implementation
  • When the current conversation already contains enough context to synthesize a first PRD draft

Note: This skill works best for well-scoped, single-feature PRDs. For large epics with many sub-features, use this per sub-feature and link them together.

Before You Write This PRD (Risk Gate)

Before formalizing requirements, briefly verify that the feature is worth building. Skip this gate only when the build decision is already validated (e.g., a committed roadmap item, a contractual requirement, or a P0 bug fix).

Feature Demand Level Check

LevelSignalAction
0Founder/stakeholder anxiety — "competitors have it" or "feels incomplete"Defer; validate first
1One user request, no behavior proofValidate before writing a PRD
2Repeated requests from target users, no usage proof yetWrite a lightweight PRD; plan validation alongside it
3Workflow blocker — users cannot complete the core job without itWrite the PRD and build
4Revenue or retention blocker — users churn or refuse to pay without itWrite the PRD and build now

Proceed to a full PRD for Level 3 or Level 4. For Level 0–2, start with the smallest validation step (a buyer conversation, a landing page, a manual pilot) and feed the findings into the PRD's Problem Statement and JTBD sections instead of skipping straight to requirements.

Minimum Risk Check (for Level 1–2)

Answer these before writing requirements:

  • Demand: Is there evidence a specific user urgently needs this?
  • Switching reason: What makes the current alternative insufficient?
  • Distribution: Is there a repeatable way to reach the target user after launch?
  • Retention: Is there a reason users would return after the first use?

If any answer is "unknown", record it in the PRD's Open Questions section and tag it needs-validation-before-implementation rather than silently assuming a "yes".

Prerequisites

  • A feature idea, GitHub Issue, or user request to formalize
  • Access to any existing context: user research, analytics, prior PRDs, or CLAUDE.md/copilot-instructions.md

Workflow

1. Gather Context

If the current conversation, issue, or spec thread already contains enough information, draft from that context first. Do not restart discovery from zero unless the missing gaps would materially change scope.

Before writing, collect:

# What does the product currently do?
Get-Content README.md | Select-Object -First 40
if (Test-Path .github/copilot-instructions.md) {
    Get-Content .github/copilot-instructions.md | Select-Object -First 50
}

# Any existing product docs?
Get-ChildItem -Recurse -Include "*.md" | Where-Object { $_.Name -match 'product|spec|prd|requirements' }

# Related GitHub issues -- default with gh on PATH (Copilot CLI omits search_issues; e.g.):
gh search issues "feature-keyword" --repo my-org/my-app --state open
# When the MCP tools are exposed (no gh on PATH, or --add-github-mcp-tool=<name> for every GitHub MCP
# tool you need -- it replaces the default set -- or --enable-all-github-mcp-tools, which also enables write tools):
# Tool: github-mcp-server-search_issues
#   query: "feature-keyword in:title is:open"
#   owner: "my-org"
#   repo: "my-app"

2. Define the Job-to-be-Done (JTBD)

Before writing requirements, anchor in the user's actual goal:

When [situation/trigger],
I want to [motivation],
So that I can [expected outcome].

Example:

When I've finished coding a feature and want to commit, I want to automatically format my commit message to follow conventions, So that I can maintain a clean, readable git history without remembering the format.

3. Write the PRD

Use this template:

# PRD: [Feature Name]

## Overview
One paragraph: what this feature does and why it matters.

## Problem Statement
What user pain point or business opportunity does this address?
Include data or quotes if available.

## Jobs-to-be-Done
**Primary JTBD:**
> When [situation], I want to [motivation], so that [outcome].

**Secondary JTBDs (optional):**
- ...

## Goals & Success Metrics
| Goal | Metric | Target |
|------|--------|--------|
| Reduce time to commit | Time from ready-to-commit to pushed | < 30 seconds |
| Adoption | % of commits using conventional format | > 80% after 30 days |

## Non-Goals (Out of Scope)
- Explicit list of what this does NOT include
- Helps prevent scope creep

## User Stories
### Must Have (P0)
- As a [user], I can [action] so that [benefit]
- As a [user], I can [action] so that [benefit]

### Should Have (P1)
- As a [user], I can [action] so that [benefit]

### Nice to Have (P2)
- As a [user], I can [action] so that [benefit]

## Functional Requirements
### Core Behavior
1. [Requirement] — [Acceptance criterion]
2. [Requirement] — [Acceptance criterion]

### Edge Cases
- [Edge case] → [Expected behavior]
- [Edge case] → [Expected behavior]

## Technical Constraints
- Platform: [e.g., Windows + macOS + Linux]
- Existing systems to integrate with: [e.g., GitHub MCP, git CLI]
- Performance: [e.g., < 2s response time]

## Open Questions
- [ ] [Question that needs decision before implementation]
- [ ] [Question that needs design input]

## Appendix
- Related issues: #123, #456
- Inspired by: [prior art, competitor feature, user research]

4. Save and Link

# Save to docs directory or alongside issue
$featureName = "commit-workflow"
$prdPath = "docs/prd-$featureName.md"
# (Write the PRD content to file)

# Link from GitHub Issue (via comment or issue body update)
# gh issue comment 123 --body "PRD: docs/prd-$featureName.md"

If your workflow uses GitHub Issues as the system of record, link the PRD from the issue or publish the summary there so implementation and triage can find it later.

Examples

Minimal PRD (Solo Developer)

For a solo project, a PRD can be much lighter:

# PRD: Commit Workflow Skill

## What & Why
Add a skill that guides conventional commit message creation with emoji.
Developers lose time formatting commits manually; this automates the pattern.

## JTBD
When I'm ready to commit, I want auto-formatted conventional commit messages,
so that my git history is readable without manual effort.

## Must Have
- Emoji + type + scope + description format
- Atomic commit splitting guidance
- PowerShell code blocks for Windows compat

## Non-Goals
- Automatic commit (user still types `git commit -m`)
- Integration with GUI git clients

PRD from GitHub Issue

Issue #145: "Users can't tell which commits are breaking changes"

JTBD: When I'm reviewing git log, I want breaking changes visually distinct,
so that I can identify risky commits without reading every message.

Requirements:
- 💥 emoji + `!` suffix for breaking changes
- BREAKING CHANGE footer in body
- CI lint rule for enforcement

Tips

  • Start with JTBD, not features: features are solutions; jobs are the problems worth solving
  • Synthesize before interviewing again: if the conversation already answers the key questions, produce the first draft and ask only targeted follow-ups for true unknowns
  • Non-goals are as important as goals: they prevent scope creep during implementation
  • Acceptance criteria are testable: "users can X" is not; "clicking Save triggers validation and shows error within 200ms" is
  • PRD ≠ design spec: leave implementation details out; focus on what, not how
  • Link your PRD to the GitHub Issue: keeps traceability from discovery → delivery

See Also

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.