All skills
asyrafhussin avatar

/prd-writing

@eaaf04f

Step-by-step workflow for writing Product Requirements Documents. Use when creating PRDs, documenting features, writing specifications, or planning new products. Triggers on "write PRD", "create PRD", "document requirements", "feature spec", or "product requirements".

Use this Skill: https://skilld.dev/gh/asyrafhussin/agent-skills/prd-writing

This session only. Nothing lands on disk.

rulesmetric-no-vague-language.md

≈652 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Replace Vague Terms with Quantifiable Benchmarks

Impact: HIGH (Eliminates subjective interpretation of requirements)

Never use subjective words like "fast," "intuitive," "scalable," "user-friendly," or "secure" without defining what they mean in measurable terms. Vague language creates different expectations for every reader.

Incorrect

<!-- Bad: vague, subjective language -->
## Requirements
- The redirect should be fast
- The UI should be intuitive and user-friendly
- The system should be scalable
- Analytics should load quickly
- The application should be secure
- Links should be easy to create

Problems:

  • "Fast" to a PM means < 2s. To an engineer, it means < 50ms. To a user, it means "instant"
  • "Intuitive" is unmeasurable — intuitive to whom?
  • "Scalable" without a number is meaningless — 100 users? 1 million?
  • Every stakeholder has a different interpretation
  • QA cannot test "user-friendly"

Correct

<!-- Good: every vague term replaced with a number -->
## Requirements

| Vague Term | Quantified Requirement |
|-----------|----------------------|
| "Fast redirect" | Redirect response time: p95 < 100ms |
| "Intuitive UI" | New user creates first short URL within 30s without help |
| "Scalable" | Handles 1,000 concurrent redirects/second on single server |
| "Quick analytics" | Analytics page loads in < 3s for links with up to 100K clicks |
| "Secure" | All OWASP Top 10 vulnerabilities addressed; auth on all non-public routes |
| "Easy to create" | Link creation requires max 2 form fields (URL + title) |
| "Reliable" | 99.9% uptime measured monthly; zero data loss |
| "Responsive" | All pages functional on viewports 320px to 2560px |

Benefits:

  • Engineering, product, and QA all share the same expectation
  • Every requirement is testable with a pass/fail result
  • Performance budgets guide architectural decisions
  • No post-launch debates about whether the feature is "fast enough"

Why

  1. Shared understanding: Numbers mean the same thing to everyone
  2. Testable: Automated tests can verify quantified requirements
  3. Architecturally informative: "100ms redirect" rules out certain approaches
  4. Prevents scope creep: "Make it faster" becomes "we already hit the 100ms target"

Reference: Google Engineering Practices - Performance Budgets

Source: SKILL.md on GitHub

No alerts16d4 checks · Risk SAFE
  • Gen Agent Trust Hub16d

    The skill provides a structured workflow and comprehensive best practices for writing Product Requirements Documents (PRDs). It guides the agent through discovery, stakeholder alignment, and technical specification drafting. No malicious code, exfiltration patterns, or injection attempts were found.

  • Socket16d

    No alerts

  • Snyk16d

    Risk: LOW · No issues

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at eaaf04f. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Steadyupdated 6 months ago
metadata
{
  "author": "agent-skills",
  "version": "1.0.0"
}

README badge

README badge for asyrafhussin/agent-skills/prd-writing