Live survey304 models · 14 providers · 22,792 auditsAudited grill-with-docs · SAFE →What changedHow we verifyllms.txt

Agent skill · Audited from source

prd-generator

skill · ognjengt/founder-skillsMIT licenceAudited 28 Oct 2026 · source 90bc6f1424df

★ 646pushed 162 days ago
✓SAFEGrade B · 89/100 · 28 Oct 2026 · 90bc6f1
646repo stars · whole ognjengt/founder-skills

Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit 90bc6f1424df and found no filesystem, network or shell access and no hard-coded secrets. What it does: Generates professional PRD (Product Requirements Document) files optimized for AI coding tools. Takes a rough product idea, asks clarifying questions, and outputs a structured PDF ready to feed into AI coding assistants.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “prd-generator (ognjengt/founder-skills)”. Retrieved 28 Oct 2026. https://modelsatlas.com/skills/ognjengt/prd-generator

Trust audit

Audit 0.4.1 · source read 28 Oct 2026

Trust score89/100✓SAFE B
0 critical0 high0 medium0 low0 info

SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 16,667 of 17,602 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS0
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS0
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (0)

No findings at any severity. We re-audit when the source changes.

Audit 0.4.1 · source sha 90bc6f1424df · 28 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisherognjengt GitHub userOwner of the repository this skill is read from
  • Source folderognjengt/founder-skills/skills/prd-generatorThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdict90bc6f1424dfA new audit runs when the source changes
  • Repository statusNot archivedLast push 162 days ago
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Generates professional PRD (Product Requirements Document) files optimized for AI coding tools. Takes a rough product idea, asks clarifying questions, and outputs a structured PDF ready to feed into AI coding assistants.

---
name: prd-generator
description: Generates professional PRD (Product Requirements Document) files optimized for AI coding tools. Takes a rough product idea, asks clarifying questions, and outputs a structured PDF ready to feed into AI coding assistants.
---

# PRD Generator

## Purpose
Transform a rough product idea into a comprehensive, AI-ready Product Requirements Document (PDF) through targeted questions and structured output.

---

## Execution Logic

**Check $ARGUMENTS first to determine execution mode:**

### If $ARGUMENTS is empty or not provided:
Respond with:
"prd-generator loaded, describe your product idea"

Then wait for the user to provide their product concept in the next message.

### If $ARGUMENTS contains content:
Proceed immediately to Task Execution (skip the "loaded" message).

---

## Task Execution

### 1. MANDATORY: Read Reference Files FIRST
**BLOCKING REQUIREMENT — DO NOT SKIP THIS STEP**

Before doing ANYTHING else, use the Read tool to read:
- `./references/prd_template.md`

This template defines the exact structure your PRD must follow. **DO NOT PROCEED** to Step 2 until you have read this file.

### 2. Skip Business Context
**This skill intentionally DOES NOT read FOUNDER_CONTEXT.md.** PRDs are standalone documents that should contain all necessary context within them.

### 3. Analyze Initial Input
From the user's initial description, extract what's available:
- Product name or working title
- Core problem being solved
- Target users/audience
- Key features mentioned
- Technical preferences (if any)
- Constraints or requirements (if any)

### 4. Ask Clarifying Questions
**Use AskUserQuestion tool** to gather missing information. Ask up to 7 questions maximum, but fewer is better — stop as soon as you have enough to build a comprehensive PRD.

**Question Bank (priority order):**

| # | Question | Why it matters | Skip if... |
|---|----------|----------------|------------|
| 1 | Who is the primary user? What's their role and technical level? | Shapes all UX decisions and feature complexity | User persona is clearly described |
| 2 | What's the core problem this solves? What happens if users don't have this? | Defines the value proposition and success metrics | Problem statement is explicit |
| 3 | What are the 3-5 must-have features for launch (P0)? | Prevents scope creep, focuses MVP | Features are already listed with clear priority |
| 4 | What technology preferences or constraints exist? (Language, framework, hosting) | Determines technical architecture section | Tech stack is specified |
| 5 | Are there any integrations required? (Auth providers, APIs, third-party services) | Identifies dependencies and integration complexity | No external services mentioned or user says standalone |
| 6 | What does success look like? Any specific metrics to track? | Defines goals and success metrics section | Metrics or goals are already stated |
| 7 | Any design preferences or existing brand guidelines to follow? | Shapes UI/UX requirements section | Design is flexible or already described |

**Question strategy:**
- Ask 2-4 questions per batch using AskUserQuestion
- If the first batch answers provide enough detail, stop asking
- Never ask more than 7 questions total
- Group related questions when possible

### 5. Generate the PRD
Using the template structure from `./references/prd_template.md`, create a complete PRD:

1. **Fill every applicable section** from the template
2. **Be specific** — vague requirements produce vague code
3. **Write acceptance criteria** for every feature — make them testable
4. **Prioritize ruthlessly** — P0 should be 30-40% of features
5. **The "Implementation Notes for AI" section is mandatory** — this is what makes it AI-ready

### 6. Save and Convert to PDF

**Step 6a: Create output folder**
```bash
mkdir -p ./prd_outputs/[Project Name]/
```
Use the product name with spaces, e.g., `./prd_outputs/Churn Prevention Tool/`

**Step 6b: Save markdown file**
Write the PRD content to:
```
./prd_outputs/[Project Name]/[project_name]_PRD.md
```
Use snake_case for the filename, e.g., `churn_prevention_tool_PRD.md`

**Step 6c: Convert to PDF**
Run:
```bash
npx md-to-pdf "./prd_outputs/[Project Name]/[project_name]_PRD.md"
```
This creates `[project_name]_PRD.pdf` in the same folder.

### 7. Confirm Output
Tell the user:
- Where the PDF is saved (full path)
- Where the markdown source is saved
- Brief summary of what's in the PRD

---

## Writing Rules

### Core Rules
- Every feature MUST have testable acceptance criteria
- Use specific numbers, not vague terms ("loads in <2s" not "loads quickly")
- P0 features should be 30-40% of total features — if everything is P0, nothing is
- Data models must include field types and relationships
- API specs must include request/response examples

### PRD-Specific Rules
- Executive summary: 3-5 sentences maximum
- Problem statement: Must include current state, pain points, and business impact
- User personas: Maximum 3 primary personas — more creates confusion
- Tech architecture: Describe data flow in plain English — AI tools interpret this better than complex diagrams
- Implementation Notes for AI section: This is mandatory, never skip it

### Format Rules
- Use markdown headers consistently (# for title, ## for sections, ### for subsections)
- Use tables for structured data (metrics, data models, API specs)
- Use code blocks for JSON examples and technical specs
- Use checkboxes for acceptance criteria

---

## Output Format

The PRD follows the structure in `./references/prd_template.md`. Here's a condensed example:

```markdown
# TaskFlow — Product Requirements Document

**Version:** 1.0
**Date:** 2024-01-15
**Author:** PRD Generator
**Status:** Draft

## Executive Summary
TaskFlow is a task management tool for remote engineering teams...

## Problem Statement
**Current state:** Teams use disconnected tools...
**Pain points:**
1. Context switching between tools
2. No visibility into team workload
3. Async

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit 90bc6f1

Download folder ↗
git clone https://github.com/ognjengt/founder-skills
git -C founder-skills checkout 90bc6f1424df

The author documents no install command, so this uses the repository directly. Our verdict covers commit 90bc6f1424df only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. ✓SAFE B 89 / 100

    first audit · source 90bc6f1424df

Similar skills

Questions about prd-generator

What does the prd-generator skill do?
Claude skills for founders
Is prd-generator safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can prd-generator access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (90bc6f1424df), read on 2026-10-28. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.