- Home
- Skills
- Technical Documentation
- PRD Writer
Works with the AI Agents you already use
PRD Writer
From feature idea to 10-section engineering-ready PRD — with anti-goals, success metrics, and open questions.
$9
PRD Writer
Example session with this skill installed
I need a PRD for an in-app notification system. Users should be able to receive notifications when their reports are ready, when someone mentions them in a comment, and when a task is assigned to them.
- Read your context and instructions
- Compiled the prd writer
PRD: In-App Notification System
Status: Draft
Last updated: June 2026
1. Overview
A notification system that alerts users to relevant in-product events (report completion, mentions, task assignments) without requiring them to check back manually. Reduces time-to-awareness for time-sensitive actions.
2. Problem Statement
The problem
Users miss important events (completed reports, task assignments, mentions) because they have no in-product notification mechanism. They rely on email or manual polling.
Evidence
TBD — to be defined before engineering kickoff. Suggested: support ticket volume for "I didn't know X happened", DAU data showing return visit patterns.
Impact of not solving
Users continue to miss time-sensitive events, increasing reliance on email and reducing in-product engagement.
3. Goals & Success Metrics
Goals
- Reduce time between event and user action for high-priority notifications
- Increase daily active usage by surfacing relevant in-product events
Anti-goals
- This is NOT a marketing notification system (no promotional content)
- This is NOT an email notification replacement
- This is NOT a real-time chat or messaging system
Success metrics
| Metric | Baseline | Target | Timeframe |
|---|---|---|---|
| Notification open rate | — (new) | >40% | 30 days post-launch |
| Time-to-action on task assignments | TBD | -30% | 60 days post-launch |
[... continues for all 10 sections ...]
Connects securely to your tools. The creator never sees your data.
About this skill
A Claude Code skill that produces complete Product Requirements Documents. Gathers problem statement, goals with measurable metrics, user use cases, solution overview, scope (in and out), constraints, risks, and open questions. Enforces anti-goals. Flags all assumptions explicitly. Install SKILL.md in your Claude Code project. Triggers on: "write a PRD", "spec this out", "I need a PRD for", "help me write requirements", "document this feature". The 10-section structure it produces: Overview · Problem Statement · Goals & Success Metrics · Users & Use Cases · Proposed Solution · Scope (in/out) · Constraints & Dependencies · Risks · Open Questions · Appendix What it enforces: - Every goal must have a measurable metric - Anti-goals are required (the section most PRDs skip) - At least one risk identified - Assumptions flagged explicitly when writing from incomplete input What's included: SKILL.md + installation instructions.
The skill will not invent evidence, metrics baselines, or stakeholder names — if these are missing, it marks them as "TBD — to be defined before engineering kickoff". For very early-stage ideas with minimal context, the PRD will have more open questions than answers; this is by design. Does not generate wireframes or visual artifacts.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 5 months ago
Permissions
No special permissions declared or detected