mmaideveloper/aiarchitecttoolkit / skills/idea-task
idea-task
VerifiedInteractive idea shaping skill that turns a rough idea into a validated concept draft before prepare-task formalizes it into a Ready implementation task. Use when the user asks for early brainstorming, architecture/risk review, task decomposition, technical debt discovery, or asks questions such as '/idea-task [description]'.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.6 KB |
| agents/openai.yaml | config | 299 B |
| references/idea-draft-template.md | markdown | 902 B |
SKILL.md
View on GitHub---
name: idea-task
description: Interactive idea shaping skill that turns a rough idea into a validated concept draft before prepare-task formalizes it into a Ready implementation task. Use when the user asks for early brainstorming, architecture/risk review, task decomposition, technical debt discovery, or asks questions such as '/idea-task [description]'.
---
# Idea Task
## Purpose
Use this skill before `prepare-task` when an idea is still ambiguous.
`idea-task` does discovery and risk shaping.
`prepare-task` does implementation-ready formalization and GitHub project readiness.
## Invocation
- `$idea-task`
- `/idea-task [description]`
- `/idea-task -url "https://example.com/tasks/<id>"`
- `I have an idea, help me shape it first`
Works in VS Code, Codex Web, and Codex Desktop repository chat.
## Workflow
1. Determine source: free-text idea or `-url` existing issue/task, then restate the idea in one paragraph.
2. Review repository context relevant to the idea.
3. Read the project profile and run an applicability pre-check for privacy, AI, security, safety, accessibility, and sector governance before suggesting implementation details.
4. Ask up to three focused questions at a time.
5. Maintain an `Open Questions` list and close items as answered.
6. Produce an `Idea Draft` with:
- problem and outcome
- in-scope/out-of-scope
- likely modules/files
- risks and technical debt hotspots
- observability, logging, and rollback expectations
- compliance notes
7. Classify the next route:
- ordinary/local change: `$prepare-task`
- new or changed user/business behavior: `$generate-use-case`
- material business choice: `$generate-bdr`
- cross-component, data-flow, deployment, or quality-attribute change: `$architecture-change`
8. Decide status:
- `Ready for prepare-task`
- `Blocked` (with reasons)
9. Emit exact marker line for downstream automations:
- `Idea Task Status: Ready for prepare-task.` when ready.
10. If ready, recommend the classified next route instead of always defaulting to `$prepare-task`.
## Question Priorities
Ask only what is needed to remove ambiguity:
- user outcome and success metric
- impacted channels (API, chat simulator, mobile, web)
- data classes, retention/deletion, consent/transparency
- external API dependencies and fallback behavior
- acceptance test shape and release constraints
## Output Template
Use `references/idea-draft-template.md`.
## Guardrails
- Do not start code implementation in this skill.
- Stop and surface compliant alternatives when the proposal conflicts with applicable governance requirements.
- Keep answers concise, concrete, and directly actionable by `prepare-task`.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |