AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Design Doc

skill-works-on-your-machine-womm-skills-design-doc · by works-on-your-machine

When the user wants to create a structured design document from a raw idea, conversation, or rough notes. Also use when the user mentions 'design doc', 'spec out', 'architecture doc', 'write up the design', 'plan this out', 'brainstorm this', or 'help me think through this'. Includes a deep interview phase before writing. For breaking a design doc into epics, see epic-breakdown. For capturing sol…

No reviews yet
0 installs
38 views
0.0% view→install

Install

$ agentstack add skill-works-on-your-machine-womm-skills-design-doc

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-works-on-your-machine-womm-skills-design-doc)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
6mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Design Doc? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Design Doc

Create a structured design document from a raw idea, conversation, or rough notes. This is the first step in the build workflow — turning vague intent into something concrete enough to break into epics.

Before Starting

  1. Check golden specs access. Try to list files in ~/.womm-skills/golden-specs/. If not accessible, tell the user:
  • Add ~/.womm-skills/ to allowed paths in ~/.claude/settings.json
  • Add a reference in ~/.claude/CLAUDE.md pointing to the golden specs directory
  1. Read the project to understand stack and conventions:
  • CLAUDE.md for project-specific instructions
  • Gemfile, package.json, go.mod, Cargo.toml, etc. for language and framework
  • Existing docs/ directory for format conventions
  • Golden specs at ~/.womm-skills/golden-specs/ for context on past decisions
  1. Check for existing design docs in the project (look in docs/, docs/design/, etc.). If found, match their format and level of detail.

Process

1. Get the Idea

Ask the user to describe what they want to build — or read a conversation, notes, or rough spec they provide.

2. Deep Interview

Interview the user before writing anything. Use the following areas as a guide, but adapt to the specific idea. Questions should be specific and non-obvious — don't ask things you can infer from the project files.

Problem space:

  • What specific problem does this solve? Who has this problem?
  • Why now? What changed that makes this worth building?
  • What happens if we don't build this?

Technical implementation:

  • How does this fit into the existing system?
  • What are the key data models or state changes?
  • What external services or integrations are involved?
  • Are there concurrency, performance, or scale concerns?

UI & UX (if applicable):

  • What are the primary user flows?
  • What are the edge cases and error states?
  • What does the user see when things go wrong?

Concerns and tradeoffs:

  • What could go wrong?
  • What are we explicitly choosing not to build?
  • What's the simplest version that would be useful?

Constraints:

  • Budget, timeline, team size?
  • Existing systems we can't change?
  • Technical limits (hosting, language, framework)?

Keep interviewing until you've covered all major areas. Don't stop after 3-4 questions. Synthesize answers as you go to ask better follow-up questions.

3. Draft

Write the design doc using the template in references/design-doc-template.md. Fill every section based on what you learned from the interview and the project files.

4. Review

Present the draft to the user. Iterate based on feedback until they're satisfied.

5. Write

Save the final doc to the appropriate location:

  • Respect the project's existing docs structure
  • If no convention exists, write to docs/
  • Filename: descriptive, lowercase, hyphens (e.g., user-authentication.md)

Cross-References

  • After writing: suggest /epic-breakdown to break the design doc into implementable work
  • If the user describes a solved problem instead of a new idea: redirect to /golden-spec-create

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.