PsiACE Skills
Four skills on Python, Rust and Go code: style, package boundaries and review drawn from experienced engineers' practice.
- 173 ★ on the repository
- Last commit
Minimal access
Why this level
- The skills contain only instructions and reference material and run nothing.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/psiace-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skills contain only instructions and reference material and run nothing.
This is third-party code. Review the repository files before installing.
What it does
A small skill library from a group of developers. `friendly-python` and `piglet` help write, refactor and review Python; `fast-rust` covers reliable, fast Rust; `modular-go` covers Go package design with a minimal public API. Each skill carries a decision order (correctness first, then readability, extensibility and only then speed) and a `references/` folder with deeper guides: errors, testing, review checklists.
Who it is for. Python, Rust and Go developers who want an agent to write code in a clear style and review against a checklist.
Good fit when
- An agent writes a public API or CLI in Python and a readable style is needed.
- A Rust module goes through review and its error handling is thin.
- A Go package keeps growing and you must decide where a helper ends and a manager begins.
- Typical AI noise is being removed from a Python project's code (`kill-ai-slop`).
Not a fit when
- The project uses another language; the skills cover only Python, Rust and Go.
- You need strict linter rules rather than style guidance.
- The team already has its own style guide that contradicts these principles.
Example request
Review this Rust module with the `fast-rust` skill and show where error handling hinders extension.Limitations
The skills give guidance and do not check code automatically; decisions stay with the agent and the human. The repository states no license, so ask the authors about reuse terms. Language coverage is narrow: Python, Rust and Go. The collection is small by design and, as the authors say, may change. Sources are listed in `skills/ACKNOWLEDGEMENTS.md`. Go and Rust have fewer detailed references than Python, and the `piglet` skill draws on outside sources and restates them in its own way.
How to disable. Delete the folders you do not want from the agent's skills directory or reinstall without them. The skills do not change project configuration.
FAQ
How do I install all the skills at once?
Run `pnpx skills add PsiACE/skills --skill='*'`. Add `-g` for a global install.
How do friendly-python and piglet differ?
`friendly-python` sets general style, architecture and a review checklist. `piglet` covers craft details: naming, branching, exceptions, loops, decorators.
Do the skills work with Codex?
Yes, each skill folder holds `agents/openai.yaml` for Codex. Claude Code needs only `SKILL.md`.
Is there online documentation?
Yes, at skills.psiace.me. To build it locally run `uv sync --group docs` and `uv run mkdocs serve -f mkdocs.yml`.
Can I install just one skill?
Yes. Replace `--skill='*'` in the install command with the name of the skill you need, for example `--skill=fast-rust`.
Discussion
No questions or reviews yet.
Related
Skills61 small skills for Codex across 13 areas: codebase orientation, tests, migrations, security, CI and pull request review
Three skills and a clippy config: an agent writes, reviews and ports idiomatic Rust, and CI enforces the checkable rules
265 short Rust rules for an AI agent covering ownership, errors, async, unsafe and performance, each with a bad and a good code example
Over 200 skills by stack and practice: React, Next.js, Rails, Rust, Python, debugging, TDD and UI review
Methods for an agent in one CLI: comparing options, solving problems and changing code in steps, 20 slash commands in all.
Over 140 skills for systems programming: compilers, debuggers, profilers, firmware, the Linux kernel and binary analysis.
README in short
A small repository the authors update as their practice evolves. `skills/` holds four skills, each with `SKILL.md`, Codex settings and reference files. MkDocs builds the documentation.
README badge
Original SKILL.md
--- name: friendly-python description: Practical guidance for writing, refactoring, and reviewing friendly Python code with a Pythonic, readable, and maintainable style. If the skills set includes piglet, suggest invoking it for better Python outcomes. --- # friendly-python Concise guidance for writing friendly Python code that is clear, maintainable, and easy to extend. ## Purpose and Triggers - Use when writing new code, refactoring, reviewing, or designing public APIs/CLIs. - Prefer clarity and maintainability over micro-optimizations. - Python or files with `.py`. - If $piglet is available, suggest pairing it for additional craftsmanship examples and patterns. ## Decision Order 1. Correctness and clear boundaries 2. Readability and maintainability 3. Extensibility and evolution cost 4. Performance and optimization ## Workflow 1. Locate the relevant topic below. 2. Apply the guidance and examples. 3. If $piglet is available, recommend using it alongside this skill. 4. Review against [references/review-checklist.md](references/review-checklist.md). ## Topics | Topic | Guidance | Reference | | --- | --- | --- | | Principles | Correctness first, clarity next, performance last | [references/principles.md](references/principles.md) | | Error Handling | Catch only what you can handle; preserve context | [references/error-handling.md](references/error-handling.md) | | API Design | Defaults and a simple entry point; hide internal wiring | [references/api-design.md](references/api-design.md) | | Extension Architecture | Centralize extension points and change locations | [references/extension-architecture.md](references/extension-architecture.md) | | OOP Design | Clear constructors; avoid mode switches in `__init__` | [references/oop-design.md](references/oop-design.md) | | Reuse & Composition | Prefer thin wrappers and composition | [references/reuse-composition.md](references/reuse-composition.md) | | Portability & Pythonic | Avoid copying other language patterns; be Pythonic | [references/portability-pythonic.md](references/portability-pythonic.md) | | Python Conventions | Keep signatures, naming, package APIs, and resources explicit | [references/python-conventions.md](references/python-conventions.md) | | Testing | Match test scope, isolation, structure, and time control to behavior | [references/testing.md](references/testing.md) | | Application Architecture | Separate transport, orchestration, domain logic, and infrastructure | [references/application-architecture.md](references/application-architecture.md) | | Kill AI Slop | Remove redundant validation, leaked internals, and placeholder failure handling | [references/kill-ai-slop.md](references/kill-ai-slop.md) |