The library

learn-codebase · When applying learn-codebase patterns to specialized interface challenges.

oh-my-issues · When applying oh-my-issues patterns to specialized interface challenges.

pathfinder · Extracting duplicated concerns across features and proposing the simplest unified design before committing to a refactor.

smart-explore · Token-efficient code navigation through symbols, functions, and dependencies—use instead of reading full files to map structure before diving into implementation.

standup · Convening multiple worktrees/PRs as agents in a shared markdown chat to reconcile overlapping work without committing changes prematurely.

claude-code-plugin-release · Provide expert guidance at a specific stage.

get-unpublished-changes · Provide expert guidance at a specific stage.

github-triage · Provide expert guidance at a specific stage.

opencode-qa · When you need to prove a specific hook fired or a CLI command works without manual testing overhead.

pre-publish-review · When you're about to publish an npm package and need multi-agent coverage to catch edge cases across core, opencode, and codex layers.

publish · When you need end-to-end release orchestration across npm, GitHub, lazycodex, and Discord in a single coordinated workflow.

ent-prompt-troubleshoot · When AI output is disappointing and the fix is in the prompt, not the model — saves time vs trying new models.

remove-deadcode · When you need to clean dead code at scale without manually hunting — LSP verification prevents false positives.

security-research · When you need multi-agent security review that catches adversarial thinking across code, infra, and supply chain.

work-with-pr · When you need structured PR review that checks correctness, tests, and merge conflicts before the merge button.

lsp · When you need precise symbol tracking (dead code, refactor impact, dependency discovery) without regex fragility.

rules · Automating per-project instruction injection without rewriting CLAUDE.md for every session.

ai-firstify · Systematically aligning existing projects to AI-first principles when piecemeal refactoring has plateaued.

debugging · Production bugs and flaky tests when you need runtime truth instead of plausible code stories.

git-master · Atomic commits and complex history investigations when you need evidence-led git operations.

lcx-contribute-bug-fix · Open-source contributions when you want an end-to-end workflow from bug report to merged PR.

lcx-report-bug · Reporting bugs when you want the maintainers to reproduce quickly without back-and-forth.

programming · When building new code in Python/Rust/TypeScript/Go with strict discipline.

refactor · When you need to restructure code with full codebase awareness and risk mapping.

remove-ai-slops · When AI-generated code has stylistic/structural debt that must be cleaned before merge.

review-work · When you need comprehensive post-implementation verification across goals, QA, code, and security.

visual-qa · When you must rigorously verify UI changes before shipping.

3d-web-experience · When you need to add 3D/immersive elements to a web product.

init-deep · Generating comprehensive, hierarchical knowledge bases for agent systems with parallel discovery.

accessibility · Implementing WCAG 2.2 accessible components consistently across Web, iOS, and Android.

accesslint-audit · Systematically finding and fixing WCAG violations across code or pages with verify-after-fix loop.

accesslint-diff · Proving a UI change introduces no new accessibility regressions before merging to main.

refactor · Restructuring large code sections with safety guarantees via LSP, AST-Grep, and regression tests.

accesslint-scan · Auditing a live page comprehensively for all WCAG 2.2 violations with selector-grounded fix worklist.

remove-ai-slops · Stripping boilerplate, dead code, and defensive patterns from AI-generated code without changing behavior.

adding-tauri-splashscreen · Building desktop applications with native performance and Rust backend integrated with web UI.

adding-tauri-system-tray · Building desktop applications with native performance and Rust backend integrated with web UI.

agent-architecture-audit · Reviewing multi-agent system design for coupling, messaging patterns, and safety boundaries.

visual-qa · Detecting visual regressions and responsive design breakpoints across multiple screen sizes.

agent-eval · Measuring multi-agent system output quality, safety alignment, and hallucination rates.

agent-browser · Automating browser interactions or debugging agent reasoning state with transparent step-by-step trace.

agent-evaluation · Measuring multi-agent system output quality, safety alignment, and hallucination rates.

dev-browser · Exposing browser DevTools and debugging APIs to agent automation workflows.

agent-introspection-debugging · Automating browser interactions or debugging agent reasoning state with transparent step-by-step trace.

agent-memory-systems · Persisting multi-layer agent memory (short-term, long-term, skill) across conversations.

git-master · Maintaining clean commit history and enforcing conventional-commit discipline across branches.

agent-tool-builder · Designing tools when LLM description quality matters more than implementation quality.

security-research · Parallel security audits that separate real exploitability from concern with dedicated hunters and PoC engineers.

personalhub-youtube · YouTube API setup, credential validation, and search/trending usage without building custom integrations.

agentic-engineering · Engineering workflows where AI agents perform most implementation and evals enforce quality gates.