Skip to content
Back to skills

Workflow

ASecurity

Guided walkthrough of the 7-step workflow. Prompts at each step to invoke or skip. Use when starting a new feature or when unsure which step comes next.

  • 3 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added May 26, 2026
developmentgogitsecurity

Security analysis

A100/100

Scanned September 22, 2026

npx -y skills add pitimon/8-habit-ai-dev --skill workflow --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Workflow?

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

Security grade badge for Workflow
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/pitimon-workflow/badge)](https://www.skillsdirectory.com/skills/pitimon-workflow)

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: workflow
description: >
  Guided walkthrough of the 7-step workflow. Prompts at each step to invoke or skip.
  Use when starting a new feature or when unsure which step comes next.
user-invocable: true
argument-hint: "[feature or task to work through]"
allowed-tools: ["Read", "Glob", "Grep"]
prev-skill: none
next-skill: none
---

# 7-Step Workflow Guide (เดินทีละก้าว)

**All Habits** | **Anti-pattern**: Jumping straight to code without planning, reviewing, or monitoring

## Process

Walk the user through each step. For each step: explain what it does, ask whether to invoke or skip, and track progress.

### The 7 Steps

| Step | Skill          | Purpose                       | Skip if...                                          |
| ---- | -------------- | ----------------------------- | --------------------------------------------------- |
| 0    | /research      | Investigate before specifying | Requirements already clear, well-understood domain  |
| 1    | /requirements  | Define what, why, who         | Single-line fix, formatting change                  |
| 2    | /design        | Architecture decisions        | Bug fix in existing pattern, no architecture impact |
| 3    | /breakdown     | Atomic tasks                  | Single task, scope already clear                    |
| 4    | /build-brief   | Context before coding         | Familiar code, context still fresh                  |
| 5    | /review-ai     | Audit before commit           | Never skip                                          |
| 6    | /deploy-guide  | Staging first                 | No deployment involved                              |
| 7    | /monitor-setup | Observability after deploy    | Monitoring already comprehensive                    |

**When the problem statement itself is fuzzy** (before you know what to research), use `superpowers:brainstorming` from the `claude-plugins-official:superpowers` plugin — it ships a hard-gate collaborative design session with spec-doc-to-git. v2.4.1 removed our own `/brainstorm` skill in favor of that stronger equivalent (see ADR-006).

### For Each Step

1. **Announce**: "Step N: [skill] — [purpose]"
2. **Ask**: "Invoke /[skill], skip (with reason), or stop here?"
3. **If invoke**: Run the skill with the user's feature context
4. **If skip**: Record the reason — honest skipping is fine, lazy skipping is not
5. **Move to next step**

### After All Steps

Produce a completion summary:

```
## Workflow Summary
**Feature**: [name]

| Step | Skill          | Status  | Note                    |
| ---- | -------------- | ------- | ----------------------- |
| 1    | /requirements  | Done    | PRD drafted             |
| 2    | /design        | Skipped | Follows existing ADR    |
| 3    | /breakdown     | Done    | 4 tasks identified      |
| 4    | /build-brief   | Done    | Context brief ready     |
| 5    | /review-ai     | Done    | Verdict: PASS           |
| 6    | /deploy-guide  | Skipped | Local dev only          |
| 7    | /monitor-setup | Skipped | No deployment           |

**Steps completed**: 4/7 | **Skipped with reason**: 3/7
```

## When to Skip

- You already know exactly which step you need (just invoke it directly)
- Continuing a workflow already in progress — pick up where you left off

## Definition of Done

- [ ] Each step was either invoked or skipped with a documented reason
- [ ] Completion summary produced showing all 7 steps
- [ ] No step skipped without justification

## See Also (Cross-Plugin)

- **Looking for security scan orchestration**, not the 7-step development workflow? Invoke **`/security-workflow`** from the companion plugin `devsecops-ai-team` (v10.12.0+). It auto-detects project context, runs a recipe-based scan pipeline (webapp / container / IaC / quick / compliance), and chains results through compliance mapping and a security gate. Different skill, different scope — both invocable in the same session.
- _Naming note_: in devsecops v10.11.x and earlier this skill was named `/workflow`. It was renamed to `/security-workflow` in v10.12.0 to resolve the cross-plugin collision with this skill. The old `/workflow` continues to work as a deprecation stub through devsecops v10.x and is removed in v11.0.0 (see devsecops ADR-014).
- See [`docs/INTEGRATION.md`](https://github.com/pitimon/8-habit-ai-dev/blob/main/docs/INTEGRATION.md) for the full integration matrix between this plugin, `devsecops-ai-team`, and `claude-governance`.

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.