kumaran-is

60 mods across 3 repositories, 35 stars between them.

a2ui-angular

01

kumaran-is/claude-code-onboarding

Agent Claude Code

Expert A2UI (Agent-to-User Interface) renderer developer for Angular 21.x. Use for building A2UI renderers, component catalogs, agent communication services, action handlers, and streaming A2UI payloads. Examples:\n\n \nContext: A new AI-powered travel assistant needs a rich UI rendered from agent JSON…

35 2mo ago A 137 tokens original MIT

kumaran-is/claude-code-onboarding

Agent Claude Code

Agent that performs WCAG 2.1 compliance validation for Flutter applications. Specializes in Semantics widgets, focus management, and color contrast analysis. Examples:\n\n \nContext: Developer just built a new Flutter screen with interactive elements and images.\nUser: "Can you check this Flutter screen for…

35 2mo ago A 111 tokens original MIT

adk-reviewer

03

kumaran-is/claude-code-onboarding

Agent Claude Code

Reviews Google ADK 1.29.0 agent code. Checks ADK-specific patterns, tool contracts, session management, SSE streaming, pipeline output keys, callback correctness, and Gemini integration. Use after any changes to ADK agent services.

35 2mo ago A 54 tokens original MIT

agentic-ai-dev

04

kumaran-is/claude-code-onboarding

Agent Claude Code

Expert Agentic AI developer with Python 3.14, LangChain v1.2.8, LangGraph v1.0.7, and FastAPI 0.135.2. Use for building AI agents, RAG systems, graph workflows, tools, memory, guardrails, and tests. Examples:\n\n \nContext: User wants to build a RAG agent that searches and answers questions from internal…

35 2mo ago A 151 tokens original MIT

agentic-ai-reviewer

05

kumaran-is/claude-code-onboarding

Agent Claude Code

Code reviewer for Agentic AI services (Python 3.14, LangChain v1.2.8, LangGraph v1.0.7, FastAPI 0.135.2). Reviews graph correctness, safety, cost, testing, production readiness. Examples:\n\n \nContext: A LangGraph agent with tool nodes and RAG retrieval was just implemented.\nUser: "Review the AI agent code I just…

35 2mo ago A 134 tokens original MIT

ai-eval-designer

06

kumaran-is/claude-code-onboarding

Agent Claude Code

Use this agent to design a risk-tiered evaluation set for an AI feature. Trigger when the user says "design evals for", "I need an eval set", "what should I test for this AI feature", "build a golden dataset", or before launching an AI feature that doesn't have evals yet. Also trigger after an incident, to add the new…

35 2mo ago A 85 tokens original MIT

kumaran-is/claude-code-onboarding

Agent Claude Code

Use this agent when reviewing AI-related code (prompts, agent loops, RAG systems, tool definitions, output handling) for security issues — particularly prompt injection, tool-output injection, data exfiltration, tenant isolation breaks, and unsafe execution paths. Trigger after writing or modifying any code that calls…

35 2mo ago B 120 tokens original MIT

angular-spa

08

kumaran-is/claude-code-onboarding

Agent Claude Code

Expert Angular frontend developer. Use for building SPA UIs with standalone components, signals, lazy routing, TailwindCSS + daisyUI, and RxJS. Examples:\n\n \nContext: A new user profile page needs to be added to an Angular 21.x SPA.\nUser: "Add a user profile page to the Angular app."\nAssistant: "I'll use the…

35 2mo ago A 111 tokens original MIT

architect

09

kumaran-is/claude-code-onboarding

Agent Claude Code

Solution architect for full-stack systems. Use for designing system architecture, API contracts, sequence diagrams, deployment strategies, and tech stack decisions. Examples:\n\n \nContext: A new notification system needs to be designed before implementation begins.\nUser: "Design the architecture for our push…

35 2mo ago A 98 tokens original MIT

browser-testing

10

kumaran-is/claude-code-onboarding

Agent Claude Code

Browser automation and testing specialist. Uses playwright-cli (stateful Bash CLI) for deterministic scripted tests — network inspection, console monitoring, screenshots, tracing — and Browser-Use MCP for autonomous agent flows (goal-driven, no scripting). Use for login flows, E2E journeys, performance analysis, and…

35 2mo ago A 136 tokens original MIT

code-reviewer

11

kumaran-is/claude-code-onboarding

Agent Claude Code

Expert code review specialist. Proactively reviews code for quality, security, and maintainability. Use immediately after writing or modifying code. MUST BE USED for all code changes. Examples:\n\n \nContext: A new service and controller were just implemented in a mixed-stack project.\nUser: "Review the code changes…

35 2mo ago A 103 tokens original MIT

code-simplifier

12

kumaran-is/claude-code-onboarding

Agent Claude Code

Reviews code for unnecessary complexity, excessive nesting, redundant abstractions, over-engineering, and opportunities to reduce cognitive load without changing behavior. Complements code-reviewer (which checks correctness) by focusing purely on simplicity and readability. Use before PRs or after implementations.…

35 2mo ago A 213 tokens original MIT

comment-analyzer

13

kumaran-is/claude-code-onboarding

Agent Claude Code

Reviews code comments for accuracy against actual implementation, detects comment rot, outdated documentation, and misleading inline comments. Use proactively after writing code with docstrings, JSDoc, JavaDoc, or inline comments. Use before creating PRs to ensure documentation matches code. Examples:\n\n \nContext: A…

35 2mo ago A 194 tokens original MIT

database-designer

14

kumaran-is/claude-code-onboarding

Agent Claude Code

Database architect for PostgreSQL and Firebase Firestore. Use for schema design, migrations, ERD creation, indexing strategy, and data modeling. Examples:\n\n \nContext: A new invoicing module needs a complete database schema before implementation starts.\nUser: "Design the database schema for our invoicing and…

35 2mo ago A 106 tokens original MIT

dedup-code-agent

15

kumaran-is/claude-code-onboarding

Agent Claude Code

Detects code duplication, unused code, and dependency bloat. Specializes in technical debt identification and cleanup in Flutter + Firebase codebases. Examples:\n\n \nContext: The codebase has grown and the team suspects duplicated utility functions across modules.\nUser: "I think we have a lot of duplicated helper…

35 2mo ago A 114 tokens original MIT

deployment-engineer

16

kumaran-is/claude-code-onboarding

Agent Claude Code

Expert deployment engineer specializing in CI/CD pipelines, Docker containers, Google Cloud Run, Flutter mobile distribution, and Firebase deployments. Masters GitHub Actions workflows, progressive delivery via Cloud Run traffic splitting, container security scanning, and Workload Identity Federation. Use PROACTIVELY…

35 2mo ago A 246 tokens original MIT

kumaran-is/claude-code-onboarding

Agent Claude Code

Specialist reviewer for changes that touch shared UI component libraries (Flutter or Angular). Verifies Atomic Design tier classification, slot pattern compliance, smart/dumb adherence, and decision-tree alignment. Dispatch when a PR adds or modifies any file under the project's shared component directories.…

35 2mo ago A 201 tokens original MIT

doc-updater

18

kumaran-is/claude-code-onboarding

Agent Claude Code

Lightweight documentation updater. Use when generating or refreshing docs/CODEMAPS/ architecture snapshots after significant code changes. Integrates with /update-codemaps command. Haiku model for cost efficiency.

35 2mo ago A 44 tokens original MIT

dx-optimizer

19

kumaran-is/claude-code-onboarding

Agent Claude Code

Developer Experience specialist. Improves tooling, setup automation, and developer workflows. Use when onboarding feels slow, tasks are repetitive, or the local dev environment needs standardization. Covers environment setup verification, workflow automation, git hooks, IDE config, and README improvements. NOT for…

35 2mo ago A 68 tokens original MIT

error-detective

20

kumaran-is/claude-code-onboarding

Agent Claude Code

Runtime error investigation specialist. Analyzes logs, stack traces, and production errors after they occur. Use when diagnosing failures in running systems — parsing logs with regex, correlating errors across services, and identifying root cause from runtime data. NOT for code review (use silent-failure-hunter to…

35 2mo ago A 80 tokens original MIT

eval-reviewer

21

kumaran-is/claude-code-onboarding

Agent Claude Code

Reviews eval code after implementation. Dispatched after writing any eval tests, golden dataset cases, CI gate configuration, PromptRegistry code, or eval infrastructure. Checks: Context7-verified API usage (no memory-generated metric names), LLM judge configured correctly (not a different provider than your project…

35 2mo ago A 156 tokens original MIT

flutter-genui

22

kumaran-is/claude-code-onboarding

Agent Claude Code

Expert Flutter GenUI (A2UI) developer for conversational AI-driven UIs. Use for building GenUI widget catalogs, SurfaceController renderers, Conversation orchestration, A2UI transport adapters, declarative function evaluators, and SSE-streamed agent-to-widget flows in Flutter. Examples:\n\n \nContext: A Flutter app…

35 2mo ago A 234 tokens original MIT

flutter-mobile

23

kumaran-is/claude-code-onboarding

Agent Claude Code

Expert Flutter mobile developer. Use for building cross-platform apps with Riverpod, clean architecture, Firebase integration, and adaptive UI. Examples:\n\n \nContext: A new notification preferences screen needs to be built in the Flutter app.\nUser: "Add a notification preferences screen to the Flutter…

35 2mo ago A 105 tokens original MIT

kumaran-is/claude-code-onboarding

Agent Claude Code

Flutter mobile security and privacy compliance specialist. Use for secure storage reviews, certificate pinning, GDPR/CCPA compliance, obfuscation, and mobile-specific security hardening. Examples:\n\n \nContext: The Flutter app is approaching its first public release and needs a security review.\nUser: "Check the…

35 2mo ago A 128 tokens original MIT