lliangcol

Use this skill for backend API, DTO, or enum changes that require frontend request wrappers, types, enums, or UI adaptations.

Use this skill for durable execution records, runbooks, evidence folders, artifacts, and validation of archive layout after task completion.

Use this skill for validated local commit and optional push workflows where scope, staged content, identity, gates, and human confirmation must be handled explicitly.

Use this skill for desktop productivity UI design systems, design tokens, dense tables, status markers, theme tokens, and non-AI-looking Electron app visual standards.

Use this skill for Electron desktop packaging and release readiness across Windows, macOS, and Linux, including signing boundaries, artifacts, dry-runs, and public release checks.

Use this skill for cross-platform desktop UI QA of Electron apps, including renderer screenshots, native window behavior, shell prerequisites, task logs, and platform-specific checks.

Use this skill for database repair work that must begin with readonly confirmation, prepare minimal SQL for human DMS execution, and verify with readonly post-checks.

Use this skill for Electron desktop app implementation with secure main/preload/renderer boundaries, allowlisted IPC, local command execution, themes, i18n, and cross-platform behavior.

Generate a Java/Spring backend design plan with mandatory 12 sections and write to PLAN_OUTPUT_PATH.

Run a semi-automated Codex review -> Claude Code fix -> local checks -> Codex verify closure loop for the current branch. Use when Claude Code should orchestrate review-driven fixes with external Bash scripts, structured review artifacts, validation reports, merge readiness, and residual-risk summaries.

Run Codex-only branch review with external scripts and structured review artifacts. Use when Claude Code should collect current branch diff context and ask Codex to produce a high-confidence review report without modifying code.

Run Codex verification against an existing review.md and the current branch state. Use when Claude Code or a human has already applied fixes and only needs a focused verify pass, merge recommendation, and residual blockers.

Lightweight observer-style exploration for quick understanding alignment.

Reviewer-style exploration focused on questions and risk signals, without proposing fixes.

Unified exploration Hub Skill. Route by PERSPECTIVE to observer/reviewer style outputs; analysis only, no design or implementation.

Minimal close-out summary: what changed, why, and known limitations.

Evaluate, repair, and verify AGENTS.md quality governance for a repository. Use this skill when AGENTS.md, .Codex settings, hooks, memory, AI coding rules, or agent governance are discussed.

Evaluate, repair, and verify AGENTS.md quality governance for a repository. Use this skill when AGENTS.md, Codex settings, hooks, memory, AI coding rules, or agent governance are discussed.

Evaluate, repair, and verify CLAUDE.md quality governance for a repository. Use this skill when CLAUDE.md, .claude settings, hooks, memory, AI coding rules, or agent governance are discussed.

Use with review-fix-loop-core inside a consuming repository to apply project-specific rules, risk classification, and mechanical gates for review-fix-re-review workflows.

Use for iterative review, fix, and fresh re-review of Git worktree or branch changes. Enforces live snapshot refresh, slice invalidation, diagnostics, gate planning, and large merge coverage.

Evidence-gated, risk-tiered AI-assisted code review workflow compatible with Codex and Claude Code. Use when reviewing a PR/current branch/diff, reviewing agent-generated code, validating review findings, triaging batches of PRs, checking merge readiness, auditing high-volume agent review loops, calibrating AI reviewers, or reviewing AI-generated code. Defaults to review-only. Emphasizes reviewability intake, Not reviewable verdicts, cheap review-effort circuit breakers, risk tiering, test-change scrutiny, heterogeneous AI review evidence, CI/gate integrity, LLM/prompt security, optional diff measurement, optional fresh re-review via review-fix-loop, and human merge ownership.