Skip to content
Agentmart
Submit a tool

factory

Start or run a factory from a markdown settings file.

Skill by douglasjarquin

IndexedMIT
Write a review

Chalk Labs

Need help deploying factory?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

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/douglasjarquin-factory

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

Start or run a factory from a markdown settings file. Use when the user asks to start a factory, run one, or say where a factory lives.

Use cases: documents

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 and code. Checked Nov 1, 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/factory

What's inside

Skill name
factory
Description
Start or run a factory from a markdown settings file. Use when the user asks to start a factory, run one, or say where a factory lives.
License
MIT
Instructions
57 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Factory

## When to use it

The user asked to start a factory, run a factory, or tell you where one lives.
Software and copy use this same skill.
The difference is a markdown settings file.

## Required inputs and access

The Factory settings in the Sum clone at `skills/sum-factory/factories/<name>.md`.
A confirmed copy, when one exists, at `/workspace/sum/factories/<name>.md`.
Notion, when the floor is a Notion page.
Spiral, when the confirmed settings say to call it.
The Dispatch and Persist skills, once a run is actually approved.

## Sequence of work

Ask one question at a time. Save the answer under `/workspace/sum/factories/` before you ask the next. Recommend from a file or a page you read. Do not invent lanes or tools.

1. Name the factory. `copy` and `software` already have recommendation files.
2. Where does it live? Text files, or a Notion page? If they name a page, fetch it.
3. What are the lanes, in order? On Notion, read the status options and say the names you saw. Ask them to confirm that order.
4. Which lanes does a pass pick up, and what should happen there? Ask which lanes are theirs alone.

For copy, read `skills/sum-factory/factories/copy.md` in the clone and offer it as the recommendation. Still ask. If the Notion page shows different lanes, say what you saw and ask which order to keep.

Write the confirmed answers to `/workspace/sum/factories/<name>.md`. Read the file back.

A run comes after that confirmation, as its own request.

- `runner: software` means follow the software settings file. That lane is the software factory on a Sum installation. Do not tick it from this chat. Dispatch a worker only when the user asked you to run it there.
- `runner: directive` means dispatch one worker. The brief is the confirmed settings file plus the piece URL or path they named. The worker does only the pickup lanes, then stops. You do not call Spiral in this chat.

On a Notion floor, a piece URL is the reliable trigger. Search does not see a status property.

## How to validate the result

The settings file exists and matches what the user confirmed.
A start did not call Spiral and did not move a piece.
A run left the coordinator chat free of the lane work.
The worker report names the piece, the command, and the status it set.

## What to return

On a start: the settings path, the lanes you saw, and the action they confirmed.
On a run: the task id, the worker, and the piece or issue it took.
Then stop.

## What requires approval

Starting a factory, confirming the settings, and each run.
A recommendation file is not approval.
A piece sitting in Ready is not approval until they ask for a run.
Do not post. Do not set Approved. The user merges. The Bot never merges.

From douglasjarquin/sum, MIT.

More from douglasjarquin/sum

Signals

Stars
0
Forks
0
Open issues
2
Last commit (this folder)
Oct 6, 2026 (26 days ago)

From GitHub (fetched Oct 24, 2026).

Details

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

More from douglasjarquin

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get factory running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying factory?

CallGet help
Cite this page

factory skill. Agent Mart. https://agentmart.dev/skills/douglasjarquin-factory (last updated 2026-10-24)