Technology-specialized subagents with MCP server integrations
/plugin marketplace add rayk/lucid-toolkit
/plugin install specialize@rayk/lucid-toolkit
Architectural design specialist that produces detailed, validated design documents from requirements. Use when translating requirements into implementation-ready designs with complete API signatures, data structures, file layouts, and dependency specifications. Validates inputs for completeness and contradictions before producing design.md output.
Expert debugging specialist using systematic root-cause analysis, defense-in-depth validation, and parallel investigation patterns. Use when facing bugs, test failures, unexpected behavior, or any issue requiring methodical diagnosis rather than guessing at fixes.
Generates Flutter/Dart code with fpdart functional patterns, Riverpod 3.0 state management, and Clean Architecture. Use when user mentions Flutter, Riverpod, fpdart, widget, provider, or asks to generate Dart code.
Flutter environment specialist for setup, diagnosis, and repair. Use when builds fail, tools are missing, CI breaks, signing fails, emulators misbehave, or environment config needs tuning.
Neo4j specialist for graph database work including AuraDB cloud instances, Cypher queries, and integration with capability-driven development workflows. MUST be used for ALL Neo4j operations including schema queries, reads, writes, and destructive operations.
Generates production-grade Python 3.12+ code with modern tooling (uv, ruff, pyright), strict typing, structured logging, and TDD. Use when user mentions Python, FastAPI, Pydantic, uv, ruff, or asks to generate Python code.
Python environment specialist for setup, diagnosis, and repair. Use when uv/pip fails, pyright errors persist, Docker builds break, CI is slow, GCP deployment fails, Jupyter kernels misbehave, or Neo4j connection issues occur.
Senior Research Analyst and Fact-Checker using Firecrawl MCP for rigorous investigation with authoritative source evaluation. Use when user needs verified research, documentation gathering, fact-checking, or exploring topics requiring credible, primary sources.
Automated code review for pull requests using multiple specialized agents with confidence-based scoring
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Interactive learning mode that requests meaningful code contributions at decision points (mimics the unshipped Learning output style)