Skip to content
Back to skills

Prompt To Spec Converter

ASecurity

Convert vague product or engineering requests into executable repo-local specs before coding. Use when a task is ambiguous, feature-sized, risky, cross-cutting, or needs acceptance criteria, non-goals, testing plan, rollout notes, or architecture impact.

  • 3 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added September 12, 2026
developmentgotestingapisecurity

Works with

  • api

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned September 12, 2026

npx -y skills add pax-k/pax-fullstack-harness --skill prompt-to-spec-converter --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Prompt To Spec Converter?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Prompt To Spec Converter
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/pax-k-prompt-to-spec-converter/badge)](https://www.skillsdirectory.com/skills/pax-k-prompt-to-spec-converter)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: prompt-to-spec-converter
description: Convert vague product or engineering requests into executable repo-local specs before coding. Use when a task is ambiguous, feature-sized, risky, cross-cutting, or needs acceptance criteria, non-goals, testing plan, rollout notes, or architecture impact.
---

# Prompt To Spec Converter

## Purpose

Turn a user request into a decision-complete spec that an agent or engineer can implement without guessing.

Use `openspec-sdd` instead when the user wants formal spec-driven development, OpenSpec artifacts, proposal/design/tasks, delta specs, or archiveable feature history.

## Inspect First

- user request and conversation context
- product docs, architecture docs, related source modules, tests, schemas, routes, and existing specs
- constraints from `AGENTS.md` and validation scripts

## Procedure

1. Separate facts from assumptions.
   - Keep requirements grounded in user text and repo facts.
   - Ask only for high-impact product intent that cannot be derived.
   - Use `$research-first-change` when the spec depends on current external APIs, package behavior, standards, or security guidance.

2. Draft the spec.
   - Problem
   - Goals
   - Non-goals
   - Requirements
   - Acceptance criteria
   - Architecture impact
   - Testing plan
   - Rollout/risk
   - Assumptions
   - For UI features, include accessible-name expectations for icon-only actions.

3. Save the spec.
   - Use `docs/specs/YYYY-MM-DD-feature-name.md`.
   - Pick a short kebab-case feature name from the request.

4. Link follow-on work.
   - Reference validation commands and affected docs.
   - Use `task-planning-and-pr-protocol` when the spec is ready for implementation planning.
   - Use `openspec-sdd` when the lightweight spec should become a formal OpenSpec change.
   - Use `feature-state-machine` when the spec becomes multi-step implementation work with WIP state and evidence.

## Validation

- Confirm acceptance criteria are testable.
- Confirm assumptions are explicit.
- Confirm affected modules and validation commands are named when known.
- For web UI specs, confirm the testing plan includes a browser-level workflow when a complete user lifecycle changes.

## Completion Criteria

- The implementation task has clear goals, boundaries, tests, and risk notes.
- The spec avoids hidden product decisions.

Files in this skill

  • SKILL.md2.3 KB
  • agents/openai.yaml209 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.