Skip to content
Automation
Skill

/code

Implement ONE pending Phase from subtask plan. Only write code, do not mark checkboxes or run tests. Called by autoworker:dispatch. Ends by calling autoworker:checkpoint.

From plugin
autoworker
1811 skills1 command

Claude Code

$ npx -y skills add phj128/autoworker --skill code --agent claude-code

SKILL.md

code.SKILL.md
name: code
description: |
  Implement ONE pending Phase from subtask plan. Only write code, do not mark checkboxes or run tests.
  Called by autoworker:dispatch. Ends by calling autoworker:checkpoint.

autoworker:code — Implement One Phase

Called by autoworker:dispatch. Does one thing: implement the first incomplete Phase from the subtask plan.

Execution Flow

1. Read Subtask

Glob `subtask_*.md` (exclude subtask_template.md) →
  0 found → stop, prompt to create subtask
  1 found → use directly (backward compatible)
  multiple → grep `status:` to filter:
    - Files without status field treated as active (backward compatible)
    - Exactly 1 active → use it
    - 0 active → list all files + status, prompt user to choose
    - >1 active → report anomaly
→ Read → extract:
- Plan section Phase list (with checkboxes)
- Find the first incomplete Phase (`- [ ]` prefixed Steps)

If subtask doesn't exist or plan is empty → **stop, prompt to complete autoworker:subtask-plan first**. If all Phases are complete → **stop, prompt that autoworker:dispatch should route to autoworker:test**.

2. Implement Code

Read all Steps in the Phase, implement one by one:

1. Read Step description and target file 2. Read target file (if modifying) 3. Edit/Write to implement the code change 4. After completing the Step, continue to next Step

**Executing commands**: Steps may involve not just writing code but also executing commands (starting training, running scripts, etc.). Long-running commands use `run_in_background=true` — execute autonomously, do not ask the user to run manually. After starting background tasks like training, continue to the next step without waiting for completion.

**Forbidden**:

  • Writing code from memory without reading subtask
  • Skipping Steps (unless subtask explicitly marks as skippable)
  • Checking off boxes yourself (autoworker:checkpoint's responsibility)
  • Running tests yourself (autoworker:test's responsibility)
  • Making changes beyond the current Phase's scope
  • Asking user to manually execute commands that can be done autonomously

3. Output Summary

Phase X code implementation complete:
- Step X.1: <brief change description>
- Step X.2: <brief change description>
→ Invoking autoworker:checkpoint

4. Chain: Immediately Invoke autoworker:checkpoint

**After outputting the summary, immediately invoke `autoworker:checkpoint`. Do not wait for user instructions, do nothing else.**

Key Constraints

  • **Only one Phase**: Complete it and hand off to checkpoint — don't determine how many Phases remain
  • **Only write code**: Don't check boxes, don't run tests, don't make routing decisions
  • **Self-debug when hitting issues**: 2 consecutive failures with same approach → enter diagnostic mode (make assumptions explicit → minimum observable unit → build diagnostic tool)
  • **Don't return to user mid-way**: Unless hitting a genuine blocker requiring human decision

Important Notes

  • **Phase granularity**: A Phase typically contains 2-5 Steps, each Step is a specific file change
  • **Files in current Phase**: If a Step says "modify file X", must Read the file before Edit
  • **Chaining is mandatory**: Must invoke autoworker:checkpoint after completion, cannot skip or manually substitute
Read more
@phj128
Ships withautoworker

An auto-loop execution workflow with quality gates for Claude Code. Give Claude a task. Autoworker decomposes it, implements code, runs tests, and iterates through quality gates — autonomously looping until the job is done right.

Stars on GitHub

Momentum

stars in total
18
Star history of phj128/autoworker on GitHub, a chart by star-history.com

Other skills on autoworker.

autoworker

autoworker

Auto-loop execution workflow with quality gates. Use when starting any non-trivial implementation task. Provides automatic task…

@phj128@phj128Install Skill
checkpoint

checkpoint

Record progress to subtask: Phase completion (from autoworker:code) or test results (from autoworker:test). Auto-detects upstream…

@phj128@phj128Install Skill
deep-plan

deep-plan

Structured deep discussion for Plan Mode. Runs inside EnterPlanMode to ensure thorough questioning before writing a plan. Covers…

@phj128@phj128Install Skill
dispatch

dispatch

Read subtask.md checkbox state and route to next skill. The ONLY routing point in the execution loop. Called after…

@phj128@phj128Install Skill
gate-check

gate-check

Final quality gate before reporting task completion (Gate 3). Fills confidence assessment, runs supplementary verification for…

@phj128@phj128Install Skill
subtask-init

subtask-init

Create subtask document (first half): user confirmation, goals, assumptions. Auto-runs assumption verification experiments and…

@phj128@phj128Install Skill

More skills from other plugins.

agent-builder

agent-builder

Load immediately after an Agent intent, before the first agent_builder_* call. Governs how you create, configure, test, and…

n8n11d209k
@n8n-io@n8n-ioInstall Skill
browser-use

browser-use

Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work. Also covers…

@browser-use@browser-useInstall Skill
agent-browser

agent-browser

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling…

@vercel-labs-2@vercel-labs-2Install Skill
add-anydoc

add-anydoc

Add local office-document-to-Markdown conversion to NanoClaw agent containers with the pinned Firecrawl AnyDoc CLI. Use when…

@nanocoai@nanocoaiInstall Skill

Questions about code.

Which coding agents can use code?
code installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.