Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Claude skills based on the Layered Rails book and beyond
| Date | Stars |
|---|---|
| 2026-07-31 | 406 |
| 2026-08-06 | 411 |
Today
+5 stars today
This week
— stars this week
This month
— stars this month
Momentum
5.0
growth rate 0.00%/day
# palkan/skills
A collection of coding-agent skills by [Vladimir Dementyev](https://github.com/palkan). The bundled plugins and slash commands target [Claude Code](https://docs.anthropic.com/en/docs/claude-code), while the core skill content (`SKILL.md` + workflows + references + examples) is also usable from other environments such as Codex.
## Available Skills
### Layered Rails
Design and review Rails applications using layered architecture principles from the [Layered Design for Ruby on Rails Applications](https://www.packtpub.com/en-us/product/layered-design-for-ruby-on-rails-applications-9781806114221) book.
**Install (recommended — full plugin with commands and sub-agents):**
```
/plugin marketplace add palkan/skills
/plugin install layered-rails@palkan-skills
```
**Commands:**
| Command | Purpose |
|---------|---------|
| `/layered-rails:analyze` | Full codebase architecture analysis |
| `/layered-rails:analyze-services` | Audit `app/services/` — conventions, clusters, layer hygiene, test consequences |
| `/layered-rails:analyze-callbacks` | Score model callbacks, find extraction candidates |
| `/layered-rails:analyze-gods` | Find god objects via churn x complexity |
| `/layered-rails:review` | Review code changes for layer violations |
| `/layered-rails:spec-test` | Run specification test on specific files |
| `/layered-rails:plan [goal]` | Plan incremental adoption of layered patterns |
**Install via [skills.sh](https://skills.sh/) (skill content only — no slash-command bindings):**
```
npx skills add palkan/skills --skill layered-rails
```
skills.sh delivers the `skills/layered-rails/` tree — `SKILL.md`, `workflows/`, `references/`, and `examples/`. The `/layered-rails:*` slash commands and the `layered-rails-planner` / `layered-rails-reviewer` sub-agent registrations are not part of the skill spec and won't be copied, but every workflow they wrap is in `workflows/` and can be invoked by name in plain language ("run the layered-rails review workflow on this diff"). Use this path when you want the same skill in Codex, the Claude API, or any other agent that supports skills.sh; use `/plugin install` when you want the slash commands.
### Integration with compound-engineering
We recommend asking Claude itself to update the instructions for Compound Engineering to include Layered Design features according to your needs. Below you can find some examples.
If you use the [compound-engineering](https://github.com/EveryInc/compound-engineering-plugin) plugin, add explicit instructions to include layered-rails planning and reviewing agents.
In your `compound-engineering.local.md` file:
```md
---
review_agents:
- layered-rails
- rails-reviewer
- security-sentinel
# - ...
---
# ...
# also worth adding something like:
We are **gradually adopting layered design principles** from "Layered Design for Ruby on Rails Applications" — clean abstraction boundaries, explicit layers, and specification tests...
```
Similarly, for planning features, add to your `CLAUDE.md` (or `AGENTS.md`, or whatever) smth like:
```md
# ...
### For planning agents
When planning new features or architectural changes, use the `layered-rails` skill for analysis:
- `/layered-rails:plan` — plan incremental adoption of layered patterns
- `/layered-rails:analyze` — full codebase architecture analysis
- `/layered-rails:review` — review code from a layered architecture perspective
- `/layered-rails:spec-test` — apply the specification test to evaluate layer placement
```
Alternatively, you can try to add shorter instructions to `CLAUDE.md`:
```md
Extend the list of **review agents** with the `layered-rails:layered-rails-reviewer` agent to check for architecture layer violations. Must be applicable to such commands from the compound-engineering plugin as `/workflow:review`, `/plan_review`, and similar.
Extend the list of **planning agents** with the `layered-rails:layered-rails-planner` agent to plan refactoring according to the layerExcerpt of 4,166 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:0297b4e17708b7f8, llm:description: 'Claude skills based on the Layered Rails book and beyond' (Python repository)
matched fp:0297b4e17708b7f8, llm:description: 'Claude skills based on the Layered Rails book and beyond' (Python repository)
matched fp:0297b4e17708b7f8, llm:description: 'Claude skills based on the Layered Rails book and beyond' (Python repository)