AI CLI configuration audit and optimization agent. Collects official best practices from the web for Codex CLI (~/.codex/), Gemini CLI (~/.gemini/), and Claude Code (~/.claude/), analyzes config.toml/settings.json/CLAUDE.md/GEMINI.md/AGENTS.md/permissions/commands/hooks/rules/MCP/extensions, and proposes improvements in Before/After diff format. Never edits configuration files directly.

Detect deprecated libraries, suggest native API replacements, and create PoCs for technology migrations. Use for stack modernization, legacy code updates, and technology refresh.

SVG icon/illustration generation, icon system design, and sprite symbol construction. Use when vector assets are needed.

Automated code review agent orchestrating tri-engine parallel review (Codex + Gemini + Claude Code) via subagents, with grounding verification that ships only findings worth fixing. Handles PR review automation and pre-commit checks. Detects bugs, security vulnerabilities, logic errors, and intent misalignment. Complements Zen's refactoring suggestions. Use when code review or quality checks are needed.

Unified release planning, execution, and tracking. Covers versioning strategy, CHANGELOG generation, release notes, rollback plans, and feature flag design for safe, predictable delivery.

<!--

Analyzes codebases to generate and manage Figma Make Guidelines.md packages, designs prompt strategies, and validates output. Use when preparing optimized input for Figma Make.

Test data and fixture generation agent. Use when factory pattern design, boundary value data generation, synthetic data generation, or seed data management is needed.

Document format conversion (Markdown, Word, Excel, PDF, HTML). Converts specifications from Scribe and reports from Harvest into distributable formats. Also generates reusable conversion scripts.

Mobile development specialist. Implements best practices for React Native/Flutter/SwiftUI/Jetpack Compose, navigation, offline support, and store review preparation. Use when mobile app development is needed.

Browser automation agent using Playwright and Chrome DevTools to complete tasks. Automates data collection, form interaction, screenshot capture, and network monitoring. Task completion focus (vs Voyager for E2E testing). Use when browser automation is needed.

LLM-optimized folder structure design. Audits and restructures project directories for context efficiency, progressive disclosure, and prompt cache performance. Use when agents can't find files, context costs are high, or CLAUDE.md hierarchy needs organization. Don't use for general repo structure (Grove), config audit (Hone), or skill generation (Sigil).

Meta-orchestrator that coordinates specialist AI agent teams. Decomposes requests into minimum viable agent chains, spawns each agent as an independent session via Agent tool in AUTORUN modes, and drives to final output automatically. Use when multi-agent coordination is needed.

AI/ML design and evaluation specialist covering prompt engineering, RAG design, LLM application patterns, AI safety, evaluation frameworks, MLOps, and cost optimization.

Autonomous loop runner for nexus-autoloop. Generates script sets from goals, designs operation contracts, audits live loops, and recovers state — delivering end-to-end runners that complete reliably.

Usability improvement, interaction quality enhancement, cognitive load reduction, feedback design, and a11y compliance. Use when improving UX usability or interaction feel.

Deep GitHub Actions workflow expert. Covers trigger strategy, security hardening, performance optimization, PR automation, and Reusable Workflow design. Use when new GHA workflow design or advanced optimization is needed.

Faithful reproduction agent that generates pixel-accurate HTML/CSS code from image mockups (PNG/JPG/screenshots) and performs visual verification. Use when mockup-to-code generation is needed.

Internationalization (i18n) and localization (l10n) specialist. Hardcoded string extraction to t() functions, Intl API integration for date/currency/number formatting, translation key structure management, and RTL layout support. Use when multi-language support or i18n setup is needed.

OWASP ZAP/Burp Suite/Nuclei integration, penetration test planning, DAST execution, and vulnerability scanning. Use when dynamic security testing, penetration testing, or runtime vulnerability validation is needed. Complements Sentinel static analysis.

UX writing specialist for user-facing text. Handles microcopy, error messages, voice and tone design, onboarding copy, and accessibility text. Use when UX writing or content strategy is needed.

KPI definition, tracking event design, and dashboard specification. North Star Metric, funnel analysis, and cohort analysis design. GA4/Amplitude/Mixpanel/PostHog integration. Use when metrics foundation is needed.

Add JSDoc/TSDoc, update READMEs, replace any types with proper definitions, and add high-value comments to complex logic. Use when documentation gaps, unclear intent, or type safety improvements are needed.

Edge-case test addition, flaky test repair, and coverage improvement. Use when test gaps need filling, reliability needs raising, or regression tests need adding. Multi-language support (JS/TS, Python, Go, Rust, Java).

Priority quantification agent. Scores and orders competing items using ICE/RICE/WSJF/MoSCoW/Cost of Delay/Kano frameworks. Does not write code.

Retention strategy, re-engagement, and churn prevention. Retention analysis frameworks, re-engagement trigger design, gamification elements, habit formation design, and loyalty programs. Use when engagement tactics are needed.

Git history investigation, regression root cause analysis, and code archaeology specialist. Time-travels through commit history to uncover truth. Use when git history investigation or regression analysis is needed.

Narrative design agent that tells product and feature use cases as customer-centric stories. Use when customer experience storytelling, scenario stories, or product narratives are needed.

Database schema design, migration planning, and ER diagram specialist. Handles normalization, index strategies, and relation definitions. Use when DB schema design is needed.

Search engine and vector DB design specialist. Use when full-text search, vector search, or hybrid search design, index optimization, or RAG retrieval layer implementation is needed.

Static security analysis agent. Handles hardcoded secret detection, SQL injection prevention, input validation, security header configuration, and dependency CVE scanning. Use when security auditing or vulnerability fixing is needed. Don't use for runtime exploit verification (Probe), general code review (Judge), CI/CD pipeline management (Gear), or detection rule authoring (Vigil).

Storybook story creation, catalog management, and Visual Regression integration. Handles UI component documentation, visual testing, and CSF 3.0/Factories format story creation. Supports Storybook 10 (ESM-only) and React Cosmos. Use when component documentation or visual testing is needed.

Load testing, contract testing, chaos engineering, mutation testing, and resilience verification specialist. Use when system limit verification, non-functional testing, or reliability validation is needed.

Meta-tooling agent that analyzes project codebases, tech stacks, and conventions to dynamically generate Claude Code skills optimized for that project. Places skills in both .claude/skills/ and .agents/skills/ to boost dev efficiency.

AI image generation code creation using Gemini API. Handles text-to-image generation, image editing, and prompt optimization. Use when image generation code is needed.

Ghost hunter for 'invisible' concurrency, async, and resource management issues. Detects, analyzes, and reports Race Conditions, Memory Leaks, Resource Leaks, and Deadlocks. Does not write code. Delegates fixes to Builder.

Detects unnecessary files, identifies unused code, finds orphaned files, and proposes safe deletion. Use when repository cleanup, dead code removal, or project tidying is needed. Don't use for code removal execution (Builder), deletion review (Judge), repo structure (Grove), or scope cutting (Void).

EXPLAIN ANALYZE analysis, query execution plan optimization, index recommendations, and slow query detection/fixing. Use when DB performance improvement or query optimization is needed. Complements Schema's schema design. Don't use for schema design/migrations (Schema), application code rewrites (Builder), non-DB performance (Bolt), or unknown root cause investigation (Scout).

Detection Engineering agent. Designs Sigma/YARA rules, maps detection coverage, designs threat hunting hypotheses, executes Purple Team Blue side, and integrates Detection-as-Code CI/CD. Use when defensive security verification is needed.

User feedback collection, NPS survey design, review analysis, sentiment analysis, feedback classification, and insight extraction reports. Use when establishing feedback loops.

YAGNI verification, scope cutting, pruning, and complexity reduction proposals. A 'subtraction' agent that questions the justification for every code, feature, process, document, design, spec, dependency, and config. Does not write code.

E2E testing specialist. Handles Playwright/Cypress/WebdriverIO configuration, Page Object design, auth flows, parallel execution, visual regression, a11y testing, and CI integration. Validates entire user journeys. Use when E2E test creation is needed.

Variable name improvement, function extraction, magic number constants, dead code removal, and code review. Use when code is hard to read, refactoring is needed, or PR review is required. Does not change behavior. Don't use for bug/security detection (Judge), new tests/coverage (Radar), architecture analysis (Atlas), or feature implementation (Builder).

Tech blog/article series authoring for note/Zenn/Qiita/dev.to. Not for specs (Scribe) or microcopy (Prose).

>-

>-

>-

>-

An experimental prompt engineering project. Simulates a tech department management process with multiple specialists, cross review, and competitive improvement. Default 3 person team, 11 person involution mode optional. Emergency exit always available. A fun exploration of LLM capability boundaries.

AI CRO Expert — landing page audit, A/B test plan, funnel analysis, heatmap interpretation, form optimization สำหรับเว็บไซต์และ e-commerce ไทย