vibeeval

Scans the .claude/ directory for security misconfigurations, exposed secrets, and unsafe permissions.

Covers React error boundary hierarchy, fallback UI patterns, offline-first fallback, retry mechanisms, and graceful degradation.

React, Next.js, and TypeScript frontend patterns, focusing on component design, state management, and performance.

Phase transition validation with artifact checklists, PASS/FAIL/CONDITIONAL gates for workflows.

AI software team for Claude Code: 138 agents, 295 skills, 73 hooks. This self-learning, multi-agent swarm features autonomous skill evolution.

Message queue patterns with BullMQ, Kafka, RabbitMQ, covering saga, outbox, dead letter queue, and exactly-once semantics.

Agent-to-agent communication templates, covering formats for standard handoff, QA verdict (PASS/FAIL), escalation, bug reports, security findings, and status updates.

Incident severity classification, runbook templates, root cause analysis, and post-incident review.

AI software team for Claude Code, featuring 138 agents, 295 skills, and 73 hooks. It operates as a self-learning, multi-agent swarm with autonomous skill evolution.

AI software team for Claude Code - 138 agents, 295 skills, 73 hooks. Self-learning, multi-agent swarm, autonomous skill evolution.

Ensures WCAG 2.2 AA compliance, incorporating ARIA patterns, keyboard navigation, and screen reader optimization.

A terminal pet living in your statusline, featuring 12 species and 5 stats (DEBUGGING, PATIENCE, CHAOS, WISDOM, SPEED). It reacts to your workflow, happy with passing tests, sad with failed builds, and excited in swarm mode, with species determined by user ID.

AI software team for Claude Code, comprising 138 agents, 295 skills, and 73 hooks. It is a self-learning, multi-agent swarm with autonomous skill evolution.

AI software team for Claude Code: 138 agents, 295 skills, 73 hooks. A self-learning, multi-agent swarm with autonomous skill evolution.

Framer Motion patterns, page transitions, skeleton loading, scroll-linked animations, and gesture-based interactions for React.

A fully autonomous PR pipeline that manages creation, CI, error fixing, review feedback, and merge, all controlled by budget and time limits.

AI software team for Claude Code, featuring 138 agents, 295 skills, and 73 hooks. It's a self-learning, multi-agent swarm with autonomous skill evolution.

Voice framework, visual identity standards, color palette management, typography specifications, and brand consistency.

AI software team for Claude Code - 138 agents, 295 skills, 73 hooks. Self-learning, multi-agent swarm, autonomous skill evolution.

A Steam-style achievement system featuring XP, levels, streaks, and skill trees. It gamifies the development workflow with 25 achievements across 5 categories.

A framework for measuring and tracking agent response quality over time, detecting regressions before they reach production. Use it when evaluating agent changes, auditing quality, or establishing performance baselines.

AI software team for Claude Code: 138 agents, 295 skills, 73 hooks. Self-learning, multi-agent swarm, autonomous skill evolution.

AI software team for Claude Code, featuring 138 agents, 295 skills, and 73 hooks. It operates as a self-learning, multi-agent swarm with autonomous skill evolution.

Agent fault tolerance and circuit breaker pattern. Stops agents that repeatedly make errors, applies a cooldown, and switches to a fallback, preventing cascading failures and infinite retry loops.

axe-core integration, WCAG 2.2 AA checklist, keyboard navigation testing, screen reader testing, and ARIA pattern validation.

An AI software team for Claude Code, comprising 138 agents, 295 skills, and 73 hooks. It features self-learning capabilities, operates as a multi-agent swarm, and exhibits autonomous skill evolution.

AI software team for Claude Code: 138 agents, 295 skills, 73 hooks. It features self-learning, a multi-agent swarm, and autonomous skill evolution.

AI software team for Claude Code, comprising 138 agents, 295 skills, and 73 hooks. It features self-learning, a multi-agent swarm, and autonomous skill evolution.

AI software team for Claude Code, comprising 138 agents, 295 skills, and 73 hooks. This self-learning, multi-agent swarm features autonomous skill evolution.

Tests agent behavior and verifies protocol compliance, using assertion-based tests to measure adherence to defined roles. It detects personality drift, role violations, and output quality regressions.

An AI software team for Claude Code, featuring 138 agents, 295 skills, and 73 hooks. It is characterized by self-learning, a multi-agent swarm, and autonomous skill evolution.

API design, versioning, testing, schema validation, and contract testing patterns for REST and GraphQL APIs.

AI software team for Claude Code, featuring 138 agents, 295 skills, and 73 hooks. This self-learning, multi-agent swarm enables autonomous skill evolution.

AI software team for Claude Code, comprising 138 agents, 295 skills, and 73 hooks. It features self-learning, a multi-agent swarm, and autonomous skill evolution.

Post-implementation cleanup removes AI-generated bloat while preserving functionality. It runs pass-by-pass with test verification after each pass, ideal after kraken/spark complete a feature or when a codebase needs hygiene work.

AI software team for Claude Code with 138 agents, 295 skills, and 73 hooks. It features self-learning, a multi-agent swarm, and autonomous skill evolution.

An AI software team for Claude Code, comprising 138 agents, 295 skills, and 73 hooks. This self-learning, multi-agent swarm features autonomous skill evolution.

An AI software team for Claude Code, featuring 138 agents, 295 skills, and 73 hooks. It is characterized by self-learning, a multi-agent swarm, and autonomous skill evolution.

An AI software team for Claude Code, comprising 138 agents, 295 skills, and 73 hooks, features self-learning, a multi-agent swarm, and autonomous skill evolution.

AI software team for Claude Code - 138 agents, 295 skills, 73 hooks. Self-learning, multi-agent swarm, autonomous skill evolution.

AI software team for Claude Code: 138 agents, 295 skills, 73 hooks. It's a self-learning, multi-agent swarm with autonomous skill evolution.

Cloud cost optimization, token/API cost tracking, build time optimization, and FinOps patterns.

ETL/ELT patterns, batch vs streaming, idempotency, data quality framework, and pipeline orchestration.

AI software team for Claude Code: 138 agents, 295 skills, 73 hooks. It features self-learning, a multi-agent swarm, and autonomous skill evolution.

AI software team for Claude Code - 138 agents, 295 skills, 73 hooks. Self-learning, multi-agent swarm, autonomous skill evolution.

AI software team for Claude Code - 138 agents, 295 skills, 73 hooks. Self-learning, multi-agent swarm, autonomous skill evolution.

AI software team for Claude Code - 138 agents, 295 skills, 73 hooks. Self-learning, multi-agent swarm, autonomous skill evolution.

Dockerfile best practices, multi-stage builds, docker-compose, container networking, volume management, and image optimization.