code-prd

A guided workflow for turning a Product Requirements Document (PRD)—a written description of a feature—into working code. It moves through research, planning, implementation, and validation, while saving notes and progress.

In plain words
What is it for?
Use it to investigate a codebase, break a feature into tasks, implement the tasks, and check the result against the requirements.
Why use it?
It reduces the risk of coding before understanding the existing project, missing requirements, or losing track of decisions. Checkpoints help keep the implementation aligned with the PRD.

Command for Claude Code

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/yassinello/claude-plugin-prd-workflow/code-prd
Clone the repo
git clone --depth 1 https://github.com/Yassinello/claude-plugin-prd-workflow

Made for: Claude Code.

Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,740 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.

ModelPer sessionOnce invoked
Fable 5 $0.00012 $0.01740
Opus 5 $0.00006 $0.00870
Sonnet 5 $0.00002 $0.00348
Haiku 4.5 $0.00001 $0.00174

Measured 2d ago against content hash 7e02dea09726, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-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.

.claude/commands/code-prd.md · 285 lines

How it starts

The opening of the file, as written. The whole thing — 285 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code PRD Command

Guided implementation using 4-phase Context Engineering workflow.

Purpose

Bridge PRD to implementation with:

  • Context management: Never exceed 60% context for quality
  • 4-Phase workflow: Research → Plan → Implement → Validate
  • Persistent memory: Thoughts saved in .prds/thoughts/
  • Incremental progress: Clear checkpoints and validation
  • Quality assurance: Continuous alignment with PRD

Workflow

Phase 1: Research (Context Gathering)

Goal: Deep codebase understanding before writing code.

Activities:

  • Read PRD acceptance criteria and technical design
  • Explore relevant codebase areas
  • Identify existing patterns and conventions
  • Find dependencies and integration points
  • Note architectural constraints
  • Document unknowns and questions

Output: Research summary saved in .prds/thoughts/research/PRD-XXX-research.md

Key questions:

  • Where does this feature fit in the codebase?
  • What patterns/conventions should I follow?
  • What dependencies exist?
  • What could go wrong?

Phase 2: Plan (Task Breakdown)

Goal: Detailed implementation plan with tasks and estimates.

Activities:

  • Break down PRD into concrete tasks
  • Identify task dependencies
  • Estimate effort per task
  • Plan testing strategy
  • Define validation checkpoints
  • Risk mitigation planning

Output: Implementation plan saved in .prds/thoughts/plans/PRD-XXX-plan.md

Plan structure:

  • Task list with estimates
  • Dependency graph
  • Critical path identified
  • Testing approach
  • Validation criteria

Phase 3: Implement (Guided Development)

Goal: Execute plan incrementally with continuous validation.

Activities:

  • Work task-by-task from plan
  • Checkpoint after each major task
  • Validate against acceptance criteria
  • Adapt plan based on learnings
  • Document decisions and trade-offs
  • Keep context under 60%

Checkpoints:

  • After each task: Does it work? Tests pass? Aligned with PRD?
  • Before next task: Context cleanup if needed
  • Major milestones: User validation opportunity

Read the full file on GitHub · 285 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.

  1. 2d ago First seen · 285 lines · 12 tokens per session scan A 7e02dea09726

Subscribe to this mod's changes

code-prd is a command published in the GitHub repository Yassinello/claude-plugin-prd-workflow (12 stars, last pushed 9mo ago), licensed MIT. It adds 12 tokens to every session and 1,740 once invoked, about $0.0001 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.