Skip to content
Get packs

Software Factory Design Vocabulary

Gives the agent a shared vocabulary and checklist for designing AI agent pipelines that turn tickets into merged code

Source Download skill
Repository
mattpocock/skills
Stars
309,658
Licence
MIT
Files
6 files

What it does

When you ask the agent to plan an automated software factory, it uses a fixed set of terms (task, trigger, run, runner, ticket tracker, git host, orchestrator, observability platform) and steers the design toward safe patterns: start with one throttled automated trigger, keep human review on every PR, and cap retry loops. It also walks through a checklist for each task and suggests products for each service, reading the bundled survey files and then researching current options itself. It does not build or run anything by itself; it shapes the design conversation.

Great for

  • Engineers planning an AI agent system that files, triages, reviews or implements tickets and pull requests
  • Teams who want a consistent design checklist before choosing tools like GitHub Actions, Linear or a sandbox provider
  • People who want the agent to challenge overly ambitious automation designs

Skip it if

  • People looking for a tool that runs code, opens PRs or deploys anything: the skill only guides design discussion, and it can give outdated product advice because it asks the agent to research current options
  • Non-developers without an existing ticket and git workflow, since most of the vocabulary assumes one

Try asking

  • “Help me design a software factory that triages bug reports from our feedback button into Linear”
  • “Walk me through the questions I need to answer for each task in my automated PR review pipeline”
  • “Which kind of runner should I pick for an agent that implements tickets on GitHub?”
Step-by-step guide: how to use Software Factory Design VocabularyInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 6 original files, the MIT licence and an attribution note.

    software-factory.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip software-factory.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip software-factory.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Describe the specific problem you want solved (untriaged tickets, slow reviews) so the agent starts from the problem rather than the factory architecture.
  • Ask for one automated trigger at a time, with human review kept on every PR, as the skill recommends.
  • Ask the agent to verify product options against current documentation, since the bundled surveys may go stale.

What's inside

  • EXAMPLE-TASKS.md
  • GIT-HOSTS.md
  • OBSERVABILITY-PLATFORMS.md
  • RUNNERS.md
  • SKILL.md
  • TICKET-TRACKERS.md

Redistributed unmodified from mattpocock/skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-11-02. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.