Skip to content
Agentmart
Submit a tool

auto-work

Experimental: run requested Light/Dark sequences with independent review.

Skill by geldmacher

IndexedMIT★ 3
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/geldmacher-auto-work

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Experimental: run requested Light/Dark sequences with independent review.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/auto-work

What's inside

Skill name
auto-work
Description
Experimental: run requested Light/Dark sequences with independent review.
License
MIT
Instructions
19 lines in SKILL.md · view on GitHub

Files (9)

  • SKILL.md
  • agents.md · references/
  • codex.md · references/
  • cursor.md · references/
  • delivery.md · references/
  • handoff.md · references/
  • opencode.md · references/
  • operation.md · references/
  • reviewer.md · references/

SKILL.md (first lines)

# auto-work (experimental)

Before repository access, read [agreement](../../references/workflow.md), [handoffs](references/handoff.md) and supplied rules at assignment cwd. Resolve links from their file; goal/plan, scope, mode, rounds, delivery; default Light/3.

Stay here. Read [agents](references/agents.md); resolve Plan/Implement/Review/Correct with the human before phases. Prerequisites use [readiness](../../references/project-readiness.md).

Read [operation](references/operation.md); delegate Plan → Implement → fresh Review → Correct within budget → fresh Review. Continue commissioned transitions; children return, coordinator never implements/corrects. Light awaits plan approval; Dark plans within its goal. Pause consequential choices.

Valid positive Review/proof precedes Light acceptance or Dark completion. Invalid/open Review blocks both despite working code. Only commissioned delivery loads [delivery](references/delivery.md); mode grants no external/Git authority.

## Handoffs

Reuse the [dispatch template](references/handoff.md) for every child, including Correct/final Review. Validate original proof before transitions. No Review writers. Release children if supported.

## Phase records

Record selection, accepted arguments/ID and execution separately. History gaps stay unknown. Execution: `unconfirmed` without effective host proof; calls/IDs/assertions cannot confirm. Accepted calls suffice; known mismatches block.

From geldmacher/workflow, MIT.

More from geldmacher/workflow

Signals

Stars
3
Forks
0
Open issues
0
Last commit (this folder)
Oct 21, 2026 (9 days ago)
Last release
Oct 21, 2026 (8 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from geldmacher

Cite this page

auto-work skill. Agent Mart. https://agentmart.dev/skills/geldmacher-auto-work (last updated 2026-10-24)