YoungjaeDev

YoungjaeDev/my-claude-plugins

Command

Generate llms.txt from URL or local directory.

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Use when the user asks an in-depth question about a GitHub repository's internals, architecture, or implementation details — anything that benefits from AI-summarized repo documentation rather than raw code reading. Triggers on phrases like "how does handle X", "explain the architecture of ", "what's inside ", or…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Use when generating an llms.txt index file from a URL or local directory. Triggers on phrases like "generate llms.txt for ", "create an llms.txt from this docs site", or "build an llms.txt for the current repo". Follows the llms.txt standard — title, optional description, sections listing links followed by 10-15 word…

YoungjaeDev/my-claude-plugins

Plugin Claude Code

README/CHANGELOG generation with CRO best practices, plus deployment-doc templates and MOC index generation Also absorbs agent-facing authoring: write-rules (CLAUDE.md + .claude/rules generation), interview-methodology (requirements interview / grill-me stress test), tcrei-prompt (TCREI prompt rewriting), voice-prompt.

YoungjaeDev/my-claude-plugins

Command

Generate or update CHANGELOG files using Keep a Changelog format.

YoungjaeDev/my-claude-plugins

Command

Generate or restructure a client-facing deployment / procedure document with a consistent skeleton (summary + prerequisites checklist + numbered steps).

YoungjaeDev/my-claude-plugins

Command

Generate a Map of Content (MOC) index for an arbitrary docs folder — one-line hook per file plus per-domain tables, lightweight by default with an optional strict wiki style.

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

CHANGELOG format and automation reference — writing patterns based on Keep a Changelog and Conventional Commits. Loaded by the docs-forge changelog workflow.

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Deployment / procedure document skeleton — the binding structure for client-facing .md procedure docs (top summary + prerequisites checklist + numbered steps + number-match + link-don't-redefine). Loaded by the docs-forge deploy-doc workflow.

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

This skill should be used when conducting in-depth user interviews, "gathering requirements", "interview me", "ask me questions", "understand my needs before implementing", "spec-based development", when preparing comprehensive specifications before implementation, or when relentlessly stress-testing an existing plan…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Map of Content (MOC) generation spec — hook-sourcing precedence, lightweight generic vs strict wiki output, and conflict rules for indexing an arbitrary docs folder into a single MOC.md. Loaded by the docs-forge moc workflow.

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

README patterns and templates reference — writing best practices derived from awesome-readme examples. Loaded by the docs-forge readme workflow.

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Diagnose one existing skill against seven axes — frontmatter, pointer, structure, completion criteria, information hierarchy, pruning, runtime contract — and return a P0/P1/P2 finding list with concrete edits. Measures first with the bundled script, then judges only what the numbers and the body actually show. Use…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Sweep every skill in a plugin tree at once — measure the whole fleet, select a cohort from the numbers, review it across seven axes, and emit a dated audit report plus CSV with P0/P1/P2 tiers. Measurement-first so a model reads only the skills the numbers single out, except the pointer axis which covers every skill…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Write or revise a Claude Code skill so it survives both runtimes this repo ships to (Claude Code + Codex). Owns the frontmatter schema, the writing levers behind skill prose, section structure, and the packaging contract (1024-char description cap, quoted colon-space, PLUGINROOT resolver, version bump). Use when…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

TCREI prompt structuring tool. Rewrites rough prompts into Google's TCREI (Task, Context, References, Evaluate, Iterate) format so they produce consistent results when reused in the next session. Triggers: "TCREI", "structure this prompt", "prompt enhance", "make a prompt for next session", "rewrite as TCREI"…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Normalize Korean voice-mode STT input before acting on it: strip speech fillers, fix orthography and Korean-to-English code-switching, resolve garbled identifiers (file, function, branch, skill names) against the actual repo instead of guessing, and never rewrite numbers or dates. Echoes one line of what it understood…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Generate or restructure CLAUDE.md systems with modular .claude/rules/ delegation. Use when starting new projects, restructuring existing CLAUDE.md, splitting monolithic CLAUDE.md into modular rules files, modularizing project instructions, or aligning rules to Claude Code 2026 best practices. Auto-detects project…

YoungjaeDev/my-claude-plugins

Plugin Claude Code

Playwright E2E test-harness engineering — wraps Playwright's official AI test agents (npx playwright init-agents --loop=claude generates planner/generator/healer). Three skills close the planner -> generator -> healer self-improving loop: e2e-setup onboards the full harness (agents, auth separation via storageState +…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Author Playwright E2E tests for critical user flows by orchestrating the official planner and generator agents. Use when the user asks to write or add E2E tests, create a Playwright test plan, generate specs for a flow, or cover a critical user flow end-to-end. Selects CUFs (flows whose failure breaks revenue, data…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Close the Playwright self-healing loop — diagnose and repair a failing or flaky E2E run using the trace and the official healer agent. Use when the user points at a failed CI run or PR, asks to fix a broken or flaky E2E test, debug a Playwright failure, or repair the suite. Downloads the CI trace artifact, inspects it…

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Onboard a full Playwright E2E test harness in the current project — verify/install Playwright, generate the official planner/generator/healer agents via npx playwright init-agents --loop=claude, scaffold auth separation (storageState + setup project), network route mocking, an E2E operating SSOT doc, and a gated…

YoungjaeDev/my-claude-plugins

Plugin Claude Code

GitHub workflow: commit, PR, issue, worktree, unified CodeRabbit + ChatGPT-Codex cr-fix pipeline v2 (Step 5 pre-flight review detection across CR commit-status + comments + Codex reviews + Codex emoji 3-channel; autonomous Step 9 judgment replaces per-finding AskUserQuestion — LLM judges real/spurious + severity + fix.

YoungjaeDev/my-claude-plugins

Skill Claude CodeCodex

Analyze the Git changes in the files given as arguments, write a Conventional Commits message, commit, and push. Use when the user types /github-dev:commit-and-push, says "commit and push", or asks to commit specific files. Analyzes only the provided files (one logical change per commit), writes a type-prefixed…