skill-factory
AGENTS.md1 starsChanged 45 days agoMITView on GitHub
Each top-level, kebab-case directory is a standalone Codex skill package. SKILL.md defines invocation and workflow behavior; agents/openai.yaml contains user-facing metadata. create-better-spritesheet/ also owns reusable Python CLIs in scripts/ and branch-specific guidance in references/. Keep detailed material behind relative Markdown links from SKILL.md, and keep package-specific code and documentation inside that package. There is no repository-wide build system or dependency manifest. The image tools require Python 3.10+, NumPy, and Pillow. Use an isolated environment: Exercise the public production seam through spritesheet_production.py…
- Installs packages
In this repository18 other files it gives its agents.
- codebase-designSkill
- eli5Skill
- find-skillsSkill
- improve-codebase-architectureSkill
- planning-with-filesSkill
- refine-promptsSkill
- show-meSkill
- app-logo-generatorSkill
- decompose-work-atomicallySkill
- download-online-videoSkill
- download-video-transcriptsSkill
- create-better-spritesheetSkill
- video-to-spritesheetSkill
- handoff-next-goalSkill
- handoff-next-taskSkill
- human-reasoningSkill
- refine-promptsSkill
- solve-from-first-principlesSkill
The file
Files
Repository Guidelines
Project Structure & Module Organization
Each top-level, kebab-case directory is a standalone Codex skill package. SKILL.md defines invocation and workflow behavior; agents/openai.yaml contains user-facing metadata. create-better-spritesheet/ also owns reusable Python CLIs in scripts/ and branch-specific guidance in references/. Keep detailed material behind relative Markdown links from SKILL.md, and keep package-specific code and documentation inside that package.
Build, Test, and Development Commands
There is no repository-wide build system or dependency manifest. The image tools require Python 3.10+, NumPy, and Pillow. Use an isolated environment:
python3 -m venv .venv
.venv/bin/pip install numpy Pillow
.venv/bin/python create-better-spritesheet/scripts/spritesheet_production.py --help
.venv/bin/python create-better-spritesheet/scripts/spritesheet_delivery.py --helpExercise the public production seam through spritesheet_production.py advance|verify with real RGBA fixtures. Exercise evidence work through spritesheet_delivery.py diagnose|seal-delivery|verify. Treat spritesheet_pipeline.py prepare-canonical|build-package|verify-package as the compatibility and internal adapter surface: canonical authoring requests and evidence remain v3, canonical admission proofs remain v1, current pixel packages are v5, and v4 packages remain readable and verifiable for compatibility. Keep lower-level raster helpers internal to the package. Run the package regression suite with python -m unittest discover -s create-better-spritesheet/tests -p 'test_*.py' -v inside the isolated environment.
Coding Style & Naming Conventions
All in English: Every skill package must use English throughout, including SKILL.md, references/, examples, placeholders, output templates, script messages, and agents/openai.yaml. A skill's frontmatter name must match its directory. Use kebab-case for packages and reference files, snake_case for Python modules and functions, and zero-padded names for ordered frame assets. Python uses four-space indentation, type hints, pathlib.Path, focused helpers, and a main() returning an exit code. YAML uses two-space indentation.
Testing Guidelines
The spritesheet package uses Python unittest; no coverage threshold is configured. For documentation changes, verify YAML/frontmatter syntax, relative links, and package-name consistency. For Python changes, exercise --help, success paths, invalid arguments, emitted metadata, and validator failures against representative fixtures; also inspect generated sprites at native size. Place additional tests under tests/test_<module>.py and keep the isolated dependency command reproducible.
Commit & Pull Request Guidelines
Follow the repository's Conventional Commit history: docs(skill): update prompt workflow or feat(create-better-spritesheet): add atlas validation. Add an imperative bullet-list body for multi-part changes. Pull requests should state the affected skill, behavioral impact, validation commands and results, and linked issue. Include before/after images for visual pipeline changes. Keep unrelated edits separate, and avoid adding new .DS_Store, __pycache__, virtual environments, or generated outputs.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

