The library

full-artifact-report · Gate releases by running a complete, evidence-backed artifact audit that produces a signed-off HTML report documenting manifest, permissions, signatures, and size changes.

inspect-aab · Understanding the structure, module split, and native library support of an AAB before release, especially when validating Play Bundle configuration or diagnosing size anomalies.

lobster · Replacing ad-hoc agent loops with deterministic, resumable workflows that track state, gate approvals, and avoid token waste by hard-coding step sequences.

inspect-apk · Quickly understanding the surface area, dependencies, and security posture of a third-party or unverified APK before installation or distribution.

workflow · Building multi-step, asynchronous operations (onboarding, order fulfillment, polling) that can survive server restarts without consuming resources during wait periods.

check-ui-overflow · Catching layout-level regressions early during localization when translated strings are longer or right-to-left, preventing text clipping in production.

security-lens · Catching common OWASP vulnerabilities (injection, weak auth, hardcoded secrets, XSS) in code review without requiring a full security audit.

ccstatusline · Displaying live Claude Code usage, git status, and model info in a single terminal line, with persistent git caching to avoid repeated subprocess work.

git-sync · Syncing local work to remote with a clear co-authored commit message that explains WHY the change was made, not just WHAT changed.

validate-strings · Continuously validating localization correctness by checking for missing translations, broken placeholders, and XML safety issues without modifying translations.

analyze-crash · Quickly pinpointing production Android crashes by separating wrapper exceptions from root cause, mapping to lifecycle, and recommending minimal fixes.

interceptors-opentelemetry · Observing Temporal Workflow execution timing and step dependencies by emitting structured traces to OpenTelemetry-compatible backends (Jaeger, Datadog, etc.).

meteora · Swapping tokens on Solana when DLMM pools offer better pricing than constant-product AMMs.

full-review-report · Comprehensive pre-release Android review when all four specialist domains (Kotlin, lifecycle, build, resources) need parallel evaluation and consolidated reporting.

claude-mpm · Large codebases requiring multi-agent coordination across Python/TypeScript/Rust/Java/Ruby stacks with persistent session state and semantic code search.

release-check · Pre-release validation when multiple flavours, signing configs, and SDK integrations must be checked before shipping to production.

automaton · Long-running multi-step changes across Claude Code, Codex, and OpenCode when workflow phases must survive context-window resets and session restarts.

react-doctor · Quick after-change validation in React projects when a fast score + actionable fixes matter more than deep architectural review.

review-changes · Pull-request review of Android code when Kotlin correctness, lifecycle safety, and SDK integration must be jointly assessed before merge.

tina-cms · Content editing workflows when non-developers need a Git-backed UI editor for MDX blog/documentation without custom CMS build.

sdk-integration-check · SDK integration security review when third-party analytics, messaging, and deep-linking must be checked for token exposure and lifecycle safety.

create-baseline · Unattended AI development when structural phase enforcement, consent gates, and skill automation cannot be left to agent discretion.

init-project · Quick-start for new Claude Code projects needing interactive setup with preset configurations for standard, minimal, or full agent scope.

a11y-audit · Accessibility compliance review before shipping when WCAG 2.2 AA/AAA coverage and POUR (Perceivable/Operable/Understandable/Robust) must be confirmed.

kaizen-dev · K8s microservice development when standardized branching, manifest organization, and automated deployment scripts must be enforced consistently.

design-code · Component implementation across multiple frameworks when code must be production-ready, accessible, token-driven, and free of hardcoded values.

design-component · Component design specification when a new UI pattern must be formally documented before or alongside code implementation.

design-qa · Design quality automation when commit-level gates for tokens, contrast, a11y, and visual regression must prevent regressions from reaching main.

n8n-nodes-claude-sdk-agent · N8n automation workflows requiring Claude agents with MCP integrations and model-routing rules when workflow nodes must delegate to LLM-backed agents.

takt · Large structured tasks requiring plan→implement→review→fix loops when multiple agents must coordinate across isolated worktrees with human judgment gates.

treido-audit-shadcn · Enforcing architectural isolation in large shadcn/ui codebases without manual file review.

design-tokens · Generating multi-brand token systems with semantic-layer isolation and automated a11y validation.

figma-integration · Maintaining design-code sync in teams where designers use Figma and engineers build in code.

handling-deploy-cicd · Gating production deployments with environment and build verification before pushing.

governance · Preventing design-system fragmentation by enforcing versioning discipline across team changes.

openflow · Bridging the spec-to-build gap in large features by auto-translating design intent into engineer phasing.

migrate-design-system · Adopting external design systems or migrating between them without rewriting components by hand.

vkc-api-route-pattern · Standardizing API endpoint implementations across a large Next.js codebase without copy-paste drift.

performance · Hitting Core Web Vitals targets (LCP ≤ 2.5s, INP ≤ 200ms, CLS ≤ 0.1) on low-end mobile.

handling-storage-media · Serving user-uploaded media in Appwrite-backed apps without hand-rolling image optimization.

redesign · Modernizing legacy UIs to premium quality while preserving hard-won user behavior.

token-build · Maintaining a single source of truth for design tokens and auto-generating all platform outputs.

brain-monitor · Debugging memory-system behavior and verifying embedding/entity indexing in local agent projects.

tandem-agentic-engineering · Scaling engineering delivery by having teams of agents work in parallel autonomously.

ai-workflow · Bootstrapping consistent AI-assisted development workflows across teams using multiple editors.

tandem-engineering-workflow · Handling complex features that benefit from parallel specialist work with real-time status tracking.

opencode-plugin-otel · Integrating opencode sessions with observability stacks like SigNoz, Datadog, or Honeycomb

devops-scripts-authoring · Building lightweight wrapper scripts for common CI/CD tasks without heavyweight build systems

tandem-orchestration · Running multi-hour builds that spawn parallel agents and report progress back to chat without human re-entry

suggest-task-gamma · Onboarding or unblocking when you know the codebase but not what to build next