By haJ1t
Multi-agent platform: epic/subagent orchestration, squad building, multi-model debate, provider routing, agent introspection, persistent teammates, and skill marketplace. Use when orchestrating multiple agents or building agent systems.
Self-introspection, self-debug, reasoning analysis, decision trace, internal state inspection. Use when diagnosing agent behavior or unexpected reasoning.
Persistent AI teammate, cross-session memory, personality, project context continuity. Use when setting up a long-running collaborative AI partner.
Multi-squad orchestration: task decomposition, resource optimization, routing, error recovery, cross-squad dependencies. Use when planning or managing large epics.
Structured debate between multiple LLMs using Proposer→Opponent→Judge protocol for optimal solutions. Use when stress-testing a decision with adversarial reasoning.
Intelligent LLM provider selection based on task characteristics, cost-latency-quality optimization. Use when choosing the best AI provider for a task.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
107 production-grade AI agent skills across 27 installable Claude Code plugins (1 lean core
senior-dev-squadplugin with 54 skills + 26 specialty plugins with 53 skills). Complete SDLC pipeline: spec → architecture → frontend/backend → security → testing → DevOps → review → ship. Plus guardrails, orchestration, design systems, security frameworks, repo automation, MCP integration, agent platforms, marketing, product management, cross-provider teams, AI/ML engineering, compliance, distributed-systems architecture, and epic orchestration.
54 core skills + 53 specialty plugin skills = 107 skills across 27 installable plugins that turn any AI coding agent into a disciplined, production-ready development squad.
| Layer | Count | Description |
|---|---|---|
Core plugin (senior-dev-squad) | 54 skills | Cross-domain SDLC skills usable on any project |
| Specialty plugins (26) | 53 skills | Framework, database, infra, domain, product, marketing, security, AI/ML, MCP, agent platform |
| Total | 107 skills / 27 plugins |
| # | Principle |
|---|---|
| 1 | No code without spec. Agent extracts purpose, user flow, data model, edge cases, acceptance criteria first. |
| 2 | No implementation without plan. Plan breaks into atomic tasks. Each task has test + verification method. |
| 3 | Every feature proven by test. Unit, integration, e2e, a11y, security, performance — separate skills enforce each. |
| 4 | Security at every stage. Not a final review — auth, input validation, secrets, RLS, dependencies, CI, deployment — automatic gates. |
| 5 | Research before code. Compile context, search web, analyze alternatives, verify versions before writing a single line. |
| 6 | Guardrails. Context management, git hygiene, memory preservation, safety boundaries — model-invoked from their descriptions, or wired as hooks to run automatically. |
| 7 | Parallel subagents when possible. Break work into independent streams, execute concurrently, aggregate results. |
| 8 | Cross-model verification. One model writes, another reviews — catches issues a single model would miss. |
| 9 | Design is code too. Generate UIs from specs using production-grade design systems. Accessibility is non-negotiable. |
| 10 | Reviewer is ruthlessness. "It works" is not enough. Code smell, overengineering, missing tests, race conditions, security bugs, edge cases. |
| # | Skill | What It Does |
|---|---|---|
| 1 | grill | Relentlessly interviews the user one question at a time to reach shared understanding before any code is written |
| 2 | spec-first-development | Turns ideas into product specs with user stories, acceptance criteria, and edge cases |
| 3 | research-first | Compiles context, searches web, analyzes alternatives, and verifies versions before writing a single line |
| # | Skill | What It Does |
|---|---|---|
| 4 | architecture-planner | Full-stack architecture: ADRs, tradeoff tables, mermaid diagrams, CAP awareness, deployment topology |
| 5 | distributed-systems-architect | Microservices, event-driven design, data consistency, fault tolerance, multi-region, observability |
| 6 | tech-stack-advisor | Interactive tech stack selection: decision trees, tradeoff tables, stack templates |
| 7 | design-pattern-advisor | Pattern catalog with selection guide: recommends design patterns with code sketches |
| # | Skill | What It Does |
|---|---|---|
| 8 | frontend-senior-engineer | Production-grade React/Next.js: responsive, a11y, states, forms, performance |
| 9 | ui-generator | Generates production-ready UI components from specs, design system oriented |
| 10 | responsive-layout-engine | Generates responsive layouts with design system tokens for each breakpoint — mobile-first through wide screen |
| 11 | accessibility-optimizer | WCAG 2.2 AA/AAA auditing, automated fixing, and UI accessibility quality gate |
| 12 | design-system-manager | Design system catalog, selection, theming, token generation from brand, and cross-system mapping/migration |
npx claudepluginhub haj1t/senior-dev-squad-skills --plugin agent-platform-proSmart contracts, dApps, DeFi, NFT, testing, security, gas optimization. Use when building or auditing Web3 or blockchain projects.
ETL pipelines, data warehousing, streaming, orchestration, quality, governance. Use when designing or reviewing data pipelines.
LLM security: prompt injection, guardrails, PII redaction, model access control, audit. Use when securing AI apps or auditing LLM pipelines.
ML workflows, training pipelines, model evaluation, deployment, experiment tracking. Use when building or reviewing ML systems.
Supabase auth, RLS, realtime, storage, edge functions, database design. Use when building or reviewing Supabase-backed applications.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Frontend design skill for UI/UX implementation
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Memory compression system for Claude Code - persist context across sessions
Marketing skills for AI agents — conversion optimization, copywriting, SEO, paid ads, ad creative, and growth
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.