hooks-workflow

Translate rule-file hook guidance into target-appropriate hook configuration.

1|Updated Mar 4, 2026
One-click install
npx skills add https://github.com/yandy-r/claude-plugins --skill hooks-workflow
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hooks-workflow
Source: https://github.com/yandy-r/claude-plugins/tree/main/.cursor-plugin/skills/hooks-workflow
Command: npx skills add https://github.com/yandy-r/claude-plugins --skill hooks-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill turns existing rule-file hook guidance into real, target-appropriate hook configuration. It reads ycc/rules/<language>/hooks.md (plus ycc/rules/common/hooks.md when present), resolves which hook events the requested target actually supports by consulting the capability matrix, and invokes build-hook-config.sh to emit the correct artifact. Explicit per-target boundaries are enforced at every step: Claude receives a concrete JSON hooks settings fragment, Cursor receives rule-embedded advisory guidance where the matrix shows partial support, and Codex always receives an advisory-only config.toml fragment. The skill never claims parity across targets and never fabricates config for a target the matrix marks as unsupported.

Core Features & Use Cases

  • Resolve per-language hook recommendations from ycc/rules/<language>/hooks.md and ycc/rules/common/hooks.md.
  • Enforce per-target boundaries to produce Claude JSON, Cursor advisory fragments, and Codex advisory fragments.
  • Validate and verify emitted configurations with build-hook-config.sh and verify-hooks.sh to ensure parseability and compatibility.

Quick Start

Use the ycc hooks-workflow skill to generate a target-specific hooks fragment by language, target, and event.

Frequently Asked Questions about hooks-workflow

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate target-appropriate hook configuration from rule files?▼

To generate hook configuration, this skill reads language-specific rule files, resolves supported events via a capability matrix, and emits concrete JSON for Claude or advisory fragments for Cursor and Codex. It validates the output using build-hook-config.sh to ensure compatibility.

What hook config format does each target receive when translating rules?▼

Target boundaries are strictly enforced: Claude receives a concrete JSON hooks settings fragment, Cursor receives rule-embedded advisory guidance, and Codex receives an advisory-only config.toml fragment. The skill never claims parity across targets or fabricates unsupported configurations.

How do I set up the environment for translating hook guidance into target-ready configs?▼

You need python3 installed along with existing ycc/rules/<language>/hooks.md files and the common hooks.md file. The skill uses these rule files and a capability matrix to invoke build-hook-config.sh and emit verified, target-specific configurations.

Why does my hook configuration differ between Claude and Codex targets?▼

Hook configurations differ because the capability matrix enforces per-target boundaries. Claude gets executable JSON, while Codex always receives advisory-only fragments due to its partial support status. The skill never fabricates config for unsupported targets.

How are emitted hook configurations validated for parseability and compatibility?▼

Emitted hook configurations are validated using build-hook-config.sh and verify-hooks.sh. These scripts ensure the generated JSON and advisory fragments are parseable and compatible with the requested target's specific boundaries and capabilities.

Can I use this hook workflow to generate configs for targets not listed in the capability matrix?▼

No, this skill never fabricates configuration for a target the capability matrix marks as unsupported. It strictly enforces per-target boundaries, producing artifacts only for Claude, Cursor, and Codex based on their resolved support levels.