old-skills/prd/SKILL.md
Use when you have a chosen direction and need to formalize requirements into a Product Requirements Document. Use when user stories, acceptance criteria, and scope boundaries need to be written down before architecture or implementation.
npx skillsauth add olamedia/analytics-skills prdInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Create a structured Product Requirements Document with user stories, acceptance criteria, functional requirements, and scope boundaries.
When NOT to use: Requirements are already documented in a prd.md that is still accurate, or the change is a single-line fix with obvious scope.
brainstorming.md from the artifact folder (required — contains chosen direction)goal-definition.md from the artifact folder (required — contains success criteria)context-map.md from the artifact folder (recommended — for technical grounding)prd.md saved to the artifact folder (see references/formats.md for template)Read brainstorming.md, goal-definition.md, and context-map.md from the artifact folder. If brainstorming.md or goal-definition.md is missing, tell the user which skill to run first.
Also check for project documentation listed in references/context-sources.md (docs/TechStack.md, docs/ProjectStructure.md). If available, use them to inform technical considerations.
Extract:
Ask 3-5 essential clarifying questions where the chosen direction leaves gaps. Focus on:
Format questions with lettered options for quick answers:
1. How should the system handle duplicate entries?
A. Allow duplicates silently
B. Reject with a validation error
C. Warn the user and let them decide
D. Auto-merge duplicates
2. What is the scope for the first version?
A. Minimal viable — core happy path only
B. Full-featured with error handling
C. Core path + most common edge cases
D. Other: [please specify]
Users can answer with "1C, 2A" for quick iteration.
Only ask questions where the answer genuinely changes the PRD. Do not ask questions you can answer from the upstream artifacts.
For each piece of functionality, write a user story:
### US-001: [Title]
**Description:** As a [user], I want [feature] so that [benefit].
**Acceptance Criteria:**
- [ ] [Specific verifiable criterion]
- [ ] [Specific verifiable criterion]
Rules for user stories:
Number each requirement for easy reference:
- FR-1: The system must allow users to create an account with email and password
- FR-2: When a user submits an invalid email, the system must display an inline error message
- FR-3: Passwords must be at least 8 characters with one number and one special character
Be explicit and unambiguous. The reader may be a junior developer or AI agent.
Carry over the "Not Doing" list from goal definition and expand it based on the clarifying questions:
## Non-Goals (Out of Scope)
- No email verification in v1 — will be added later
- No social login — only email/password for now
- No admin dashboard — management via database directly
This section prevents scope creep. Every PRD must have non-goals.
Based on the context map, note:
Translate the success criteria from goal definition into measurable metrics:
## Success Metrics
- Registration completes in under 3 clicks
- Error rate for form submission < 5%
- Page load time < 2 seconds
Write prd.md to the artifact folder using the template from references/formats.md. Include all sections from Steps 3-7, plus an Open Questions section for anything unresolved.
Present the PRD to the user for review before saving. Apply any requested changes.
Announce the saved path:
"PRD saved to
[path]/prd.md."
The PRD reader may be a junior developer or AI agent:
| Rationalization | Reality | |---|---| | "The requirements are obvious from the brainstorming" | Brainstorming picks a direction. PRD defines every detail within that direction. Different levels of specificity. | | "User stories are too formal" | User stories force you to name the user, the action, and the benefit. That's clarity, not ceremony. | | "We'll figure out edge cases during implementation" | Edge cases found during implementation cost 10x more than edge cases found during requirements. | | "Non-goals section is unnecessary" | Without non-goals, scope creep is invisible. Every "quick addition" erodes focus. | | "Acceptance criteria slow us down" | Acceptance criteria define "done." Without them, you're done when someone feels like it. |
Before handing off, confirm:
prd.md saved to artifact folder"PRD complete. Next recommended skill: architecture — design the technical approach for these requirements."
testing
Rebase current feature branch onto master (or configured base) with automated conflict resolution. Handles pre-checks, conflict classification, and post-rebase verification. Use when the user asks to rebase, update a branch, sync with master, or resolve rebase conflicts.
development
FE feature analysis from raw idea (or YouTrack ticket) through to a split-aware developer briefs. Produces context-map.md, requirements.md, task-brief-{side}.md. Generic — project knowledge is auto-discovered.
testing
Concise technical communication that stays readable and honest. Cuts fluff about fifty to seventy percent while keeping natural flow, uncertainty markers, and human tone. Levels lite (default), mid, tight. Short SKILL body; rules below are complete.
documentation
Strip LLM tells from text so it reads human. Triggers: humanize, sounds like AI, too robotic, natural rewrite, AI slop, or obvious LLM patterns. Reference: https://en.wikipedia.org/wiki/WP:Signs_of_AI_writing