# workflow

Makes the agent plan multi-step work, push research into subagents and prove a task works before marking it done.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 2700 installs · ★ 379
- Author: brianlovin
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/brianlovin/agent-config/tree/HEAD/skills/workflow/SKILL.md
- Updated: 2026-03-25
- HTML: https://affpapa.org/en/nevibecode/skills/brianlovin-workflow

## What it does

The skill lays out six rules. For anything with three or more steps or an architectural decision, the agent enters plan mode first. If something goes sideways it stops and replans instead of pushing on. Research and parallel analysis get handed to subagents so the main context stays clean.

Every correction you make gets written to `tasks/lessons.md` as a rule the agent follows next time. The plan itself lives in `tasks/todo.md` with checkable items, and a review section is appended when the work is done. One rule is blunt: never mark a task complete without showing it works — tests, logs, a diff of behaviour before and after.

The elegance rule applies to non-trivial changes only; for simple obvious fixes the skill tells the agent to skip it rather than over-engineer. Bug reports are handled without hand-holding: read the logs and failing tests, fix the root cause. If your project has no `tasks` directory yet, you will need one, because the skill writes into it as it goes.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add brianlovin/agent-config --skill workflow -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/workflow/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/workflow/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add brianlovin/agent-config --skill workflow -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/workflow/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add brianlovin/agent-config --skill workflow -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/workflow/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install brianlovin/agent-config workflow
```

2. **Manually: put the files in place**
- File: `.github/skills/workflow/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add brianlovin/agent-config --skill workflow -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/brianlovin/agent-config --path skills/workflow
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/workflow/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add brianlovin/agent-config --skill workflow -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/brianlovin/agent-config/tree/HEAD/skills/workflow/SKILL.md

## When to use it
- turn a multi-step task into a plan before touching code
- offload codebase research to a subagent and keep context clean
- record a lesson from your correction so the mistake does not repeat
- prove a change works with tests and logs before closing it
- fix a failing CI test without being told how

## Example prompts
- Plan the auth module refactor, then implement it step by step
- Here is a bug report: check the logs and the failing test, find the cause and fix it
- Before you call this done, show me the tests passing

## Similar
- [ce-brainstorm](https://affpapa.org/en/nevibecode/skills/ce-brainstorm.md) — skill · 3900 installs
- [ce-plan](https://affpapa.org/en/nevibecode/skills/ce-plan.md) — skill · 3700 installs
- [speckit-plan](https://affpapa.org/en/nevibecode/skills/speckit-plan.md) — skill · 2500 installs
- [speckit-tasks](https://affpapa.org/en/nevibecode/skills/speckit-tasks.md) — skill · 2400 installs
- [re0-plan](https://affpapa.org/en/nevibecode/skills/re0-plan.md) — skill · 1200 installs
- [planning-with-files](https://affpapa.org/en/nevibecode/skills/zhaono1-planning-with-files.md) — skill · 1200 installs

## Other versions with the same name
- [workflow · vercel](https://affpapa.org/en/nevibecode/skills/vercel-inc-workflow.md) — 5300 installs
- [workflow · openai](https://affpapa.org/en/nevibecode/skills/workflow.md) — 18 installs

From the same set: [brianlovin/agent-config](https://affpapa.org/en/nevibecode/by/brianlovin/agent-config.md) — 12 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md