A product requirements document (PRD) is a written plan for a product or feature. This command creates a structured specification with user stories, acceptance criteria, research, and release decisions.

Use it to write a full or lightweight PRD, define user stories using INVEST criteria, add acceptance criteria, update an existing PRD, or prepare a go/no-go decision.

It turns a broad product idea into requirements that designers, developers, and stakeholders can discuss and implement. It also helps test whether the proposed work delivers clear user and business value.

write-prd — Command by yonatangross

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/yonatangross/orchestkit/write-prd

Clone the repo

git clone --depth 1 https://github.com/yonatangross/orchestkit

Per session 54 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,644 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00054	$0.01644
Opus 5	$0.00027	$0.00822
Sonnet 5	$0.00011	$0.00329
Haiku 4.5	$0.00005	$0.00164

Measured 2d ago against content hash 9f0a63dc3a28, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

write-prd scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/ork/.cursor-plugin/commands/write-prd.md · 144 lines

How it starts

The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Auto-generated from skills/write-prd/SKILL.md

Source: https://github.com/yonatangross/orchestkit

PRD — Product Requirements Document

Translate product vision and research into clear, actionable engineering specifications. Produces PRD-[product-name].md output files following an 8-section structure.

Output file naming: PRD-[product-name].md (e.g., PRD-sso-invite-flow.md)

Argument Resolution

PRODUCT = "$ARGUMENTS"  # Product name or feature, e.g., "SSO invite flow"

STEP 0: Scope Clarification

AskUserQuestion(
  questions=[{
    "question": "What type of PRD?",
    "header": "PRD Scope",
    "options": [
      {"label": "Full PRD (Recommended)", "description": "All 8 sections with research, stories, and release plan"},
      {"label": "Lightweight spec", "description": "Summary, objectives, user stories only"},
      {"label": "User stories only", "description": "INVEST stories with acceptance criteria"},
      {"label": "Update existing PRD", "description": "I have a PRD file to iterate on"}
    ],
    "multiSelect": false
  }]
)

Task Management

# 1. Create main task IMMEDIATELY
TaskCreate(subject="Write PRD: {PRODUCT}", description="8-section PRD with user stories and acceptance criteria", activeForm="Writing PRD for {PRODUCT}")

# 2. Create subtasks for each phase
TaskCreate(subject="Scope clarification", activeForm="Clarifying PRD scope")                       # id=2
TaskCreate(subject="Research and memory check", activeForm="Researching prior PRDs")               # id=3
TaskCreate(subject="Draft 8-section PRD", activeForm="Drafting PRD sections")                      # id=4
TaskCreate(subject="Write user stories and acceptance criteria", activeForm="Writing user stories") # id=5
TaskCreate(subject="Write output file", activeForm="Writing PRD file")                             # id=6

# 3. Set dependencies for sequential phases
TaskUpdate(taskId="3", addBlockedBy=["2"])  # Research needs scope first
TaskUpdate(taskId="4", addBlockedBy=["3"])  # Drafting needs research context
TaskUpdate(taskId="5", addBlockedBy=["4"])  # Stories need draft structure
TaskUpdate(taskId="6", addBlockedBy=["5"])  # Output needs all sections done

# 4. Update status as you progress
TaskUpdate(taskId="2", status="in_progress")  # When starting
TaskUpdate(taskId="2", status="completed")    # When done — repeat for each subtask

Read the full file on GitHub · 144 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 144 lines · 54 tokens per session scan A 9f0a63dc3a28

Subscribe to this mod's changes

write-prd is a command published in the GitHub repository yonatangross/orchestkit (224 stars, last pushed 2d ago), licensed MIT. It adds 54 tokens to every session and 1,644 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Data and AI Frontend Git and workflow LLM agents ai-development claude-code claude-plugin langgraph

write-prd