product-spec — community product-spec, ai_coding_project_base, community, ide skills, Claude Code, Cursor, Windsurf

v1.0.0
GitHub

About this Skill

Perfect for AI Coding Assistants needing structured product specification generation. A structured prompt framework for building software products with AI coding assistants. This toolkit guides you through product specification, technical design, and implementation planning—producing documents that AI agents can execute against.

benjaminshoemaker benjaminshoemaker
[32]
[4]
Updated: 2/24/2026

Agent Capability Analysis

The product-spec skill by benjaminshoemaker is an open-source community AI agent skill for Claude Code and other IDE workflows, helping agents execute tasks with better context, repeatability, and domain-specific guidance.

Ideal Agent Persona

Perfect for AI Coding Assistants needing structured product specification generation.

Core Value

Empowers agents to generate comprehensive product specification documents using guided Q&A sessions, producing executable documents in Markdown format, and handling deferred decisions for technical design and implementation planning.

Capabilities Granted for product-spec

Automating product specification document generation
Conducting guided Q&A sessions for project planning
Generating PRODUCT_SPEC.md files for AI-executable projects

! Prerequisites & Limits

  • Requires user input for Q&A sessions
  • Limited to generating PRODUCT_SPEC.md documents
  • Dependent on user review and refinement for accuracy
Labs Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

product-spec

Install product-spec, an AI agent skill for AI agent workflows and automation. Works with Claude Code, Cursor, and Windsurf with one-command setup.

SKILL.md
Readonly

Generate a product specification document for the current project.

Workflow

Copy this checklist and track progress:

Product Spec Progress:
- [ ] Step 1: Directory guard
- [ ] Step 2: Project root confirmation
- [ ] Step 3: Check for existing PRODUCT_SPEC.md
- [ ] Step 4: Conduct guided Q&A with user
- [ ] Step 5: Write PRODUCT_SPEC.md
- [ ] Step 6: Handle deferred decisions
- [ ] Step 7: Review and refine with user
- [ ] Step 8: Suggest next step (/technical-spec)

Directory Guard

  1. If .toolkit-marker exists in the current working directory → STOP: "You're in the toolkit repo. Run this from your project directory instead: cd ~/Projects/your-project && /product-spec"

Project Root Confirmation

Before generating any files, confirm the output location with the user:

Will write PRODUCT_SPEC.md to: {absolute path of cwd}/
Continue? (Yes / Change directory)

If the user says "Change directory", ask for the correct path and instruct them to cd there first.

Existing File Guard (Prevent Overwrite)

Before asking any questions, check whether PRODUCT_SPEC.md already exists in the current directory.

  • If it does not exist: continue normally.
  • If it exists: STOP and ask the user what to do:
    1. Backup then overwrite (recommended): read the existing file and write it to PRODUCT_SPEC.md.bak.YYYYMMDD-HHMMSS, then write the new document to PRODUCT_SPEC.md
    2. Overwrite: replace PRODUCT_SPEC.md with the new document
    3. Abort: do not write anything; suggest they rename/move the existing file first

Process

Read .claude/skills/product-spec/PROMPT.md and follow its instructions exactly:

  1. Ask the user to describe their idea
  2. Work through each question category (Problem, Users, Experience, Features, Data)
  3. Make recommendations with confidence levels
  4. Generate the final PRODUCT_SPEC.md document

Output

Write the completed specification to PRODUCT_SPEC.md in the current directory.

After writing PRODUCT_SPEC.md, verify the file exists and is non-empty by reading the first few lines. If the file was not created successfully, report the error and retry.

Deferred Requirements Capture (During Q&A)

IMPORTANT: Capture deferred requirements interactively during the Q&A process, not after.

When to Trigger

During the Q&A, watch for signals that the user is deferring a requirement:

  • "out of scope"
  • "not for MVP" / "post-MVP"
  • "v2" / "version 2" / "future"
  • "later" / "eventually"
  • "maybe" / "nice to have"
  • "we'll skip that for now"
  • "not right now"
  • "that's a separate thing"

Capture Flow

When you detect a deferral signal, immediately use AskUserQuestion:

Question: "Would you like to save this to your deferred requirements?"
Header: "Defer?"
Options:
  - "Yes, capture it" — I'll ask a few quick questions to document it
  - "No, skip" — Don't record this

If user selects "Yes, capture it":

Ask these clarifying questions (can be combined into one AskUserQuestion with multiple questions):

  1. What's being deferred? "In one sentence, what's the requirement or feature?" (Pre-fill with your understanding from context)

  2. Why defer it? Options: "Out of scope for MVP" / "Needs more research" / "V2 feature" / "Resource constraints" / "Other"

  3. Notes for later? "Any context that will help when revisiting this?" (Optional — user can skip)

Write to DEFERRED.md Immediately

After collecting answers, append to DEFERRED.md right away (don't wait until end).

If file doesn't exist, create it:

markdown
1# Deferred Requirements 2 3> Captured during specification Q&A. Review when planning future versions. 4 5## From PRODUCT_SPEC.md ({date}) 6 7| Requirement | Reason | Notes | 8|-------------|--------|-------| 9| {user's answer} | {selected reason} | {notes or "—"} |

If file exists, append:

markdown
1| {user's answer} | {selected reason} | {notes or "—"} |

(If appending to a different spec's section, add a new section header first.)

Continue Q&A

After capturing (or skipping), continue the spec Q&A where you left off. Don't break the flow.

Cross-Model Review (Automatic)

After writing PRODUCT_SPEC.md, run cross-model review if Codex CLI is available:

  1. Check if Codex CLI is installed: codex --version
  2. If available, run /codex-consult on the generated document
  3. Present any findings to the user before proceeding

Consultation invocation:

/codex-consult --research "product requirements, user stories" PRODUCT_SPEC.md

If Codex finds issues:

  • Show critical issues and recommendations
  • Ask user: "Address findings before proceeding?" (Yes/No)
  • If Yes: Apply suggested fixes
  • If No: Continue with noted issues

If Codex unavailable: Skip silently and proceed to Next Step.

Error Handling

SituationAction
PROMPT.md not found at .claude/skills/product-spec/PROMPT.mdStop and report "Skill asset missing — reinstall toolkit or run /setup"
DEFERRED.md write fails (permissions or disk)Output deferred items to terminal, warn user, continue with spec generation
Codex CLI invocation fails or times outLog the error, skip cross-model review, proceed to Next Step

Next Step

When complete, inform the user:

PRODUCT_SPEC.md created at ./PRODUCT_SPEC.md
Deferred Requirements: {count} items captured to DEFERRED.md
Cross-Model Review: PASSED | PASSED WITH NOTES | SKIPPED

Next: Run /technical-spec

FAQ & Installation Steps

These questions and steps mirror the structured data on this page for better search understanding.

? Frequently Asked Questions

What is product-spec?

Perfect for AI Coding Assistants needing structured product specification generation. A structured prompt framework for building software products with AI coding assistants. This toolkit guides you through product specification, technical design, and implementation planning—producing documents that AI agents can execute against.

How do I install product-spec?

Run the command: npx killer-skills add benjaminshoemaker/ai_coding_project_base. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for product-spec?

Key use cases include: Automating product specification document generation, Conducting guided Q&A sessions for project planning, Generating PRODUCT_SPEC.md files for AI-executable projects.

Which IDEs are compatible with product-spec?

This skill is compatible with Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer. Use the Killer-Skills CLI for universal one-command installation.

Are there any limitations for product-spec?

Requires user input for Q&A sessions. Limited to generating PRODUCT_SPEC.md documents. Dependent on user review and refinement for accuracy.

How To Install

  1. 1. Open your terminal

    Open the terminal or command line in your project directory.

  2. 2. Run the install command

    Run: npx killer-skills add benjaminshoemaker/ai_coding_project_base. The CLI will automatically detect your IDE or AI agent and configure the skill.

  3. 3. Start using the skill

    The skill is now active. Your AI agent can use product-spec immediately in the current project.

Related Skills

Looking for an alternative to product-spec or another community skill for your workflow? Explore these related open-source skills.

View All

widget-generator

Logo of f
f

f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.

149.6k
0
AI

flags

Logo of vercel
vercel

flags is a Next.js feature management skill that enables developers to efficiently add or modify framework feature flags, streamlining React application development.

138.4k
0
Browser

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
AI

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
AI