noah-sheldon/ai-dev-kit
40 subagents
Subagent
ai-judgeGatekeeper agent for structured rubric scoring of research, architecture, security, and planning outputs. Validates completeness, correctness, security, feasibility, and testability before implementatai-judge.agentai-judge.mdarchitectSystem design specialist for scalability, technical decisions, and component boundaries. Spawns parallel deep sub-agents for repository research and architecture spikes. Emits ADRs and requests AI-Judarchitect.agentarchitect.mdarchitecture-debt-analyzerAnalyzes architecture debt — circular dependencies, God modules, tight coupling, layer violations, missing separation of concerns, and structural decay. Produces a structured architecture debt report.architecture-debt-analyzer.agentarchitecture-debt-analyzer.mdbuild-error-resolverBuild and type error specialist who fixes compilation failures incrementally using Pyrefly (Python) and tsc/ESLint/Biome (TypeScript). Keeps diffs minimal, preserves behavior, and restores green CI fabuild-error-resolver.agentbuild-error-resolver.mdchrome-ext-developerWXT Chrome extension specialist for Manifest V3, content scripts, service workers, Chrome APIs, React integration, cross-browser compatibility, Chrome Web Store publishing, and extension testing/debugchrome-ext-developer.agentchrome-ext-developer.mdcode-quality-analyzerAnalyzes code quality debt — God files, long functions, deep nesting, cyclomatic complexity, naming inconsistencies, DRY violations, and readability issues. Produces a structured quality debt report wcode-quality-analyzer.agentcode-quality-analyzer.mdcode-reviewerCode quality and maintainability reviewer with absorbed quality-nonconformance playbook. Checks for bugs, anti-patterns, performance issues, test coverage gaps, and regressions across Python, TypeScricode-reviewer.agentcode-reviewer.mdcodebase-analyzerDeep codebase analysis agent. Maps architecture, data flow, coding styles, patterns, dependency graphs, and structural health. Produces a structured report after scanning. Use this to understand any ccodebase-analyzer.agentcodebase-analyzer.mdcodebase-learnerCodebase learning agent. Systematically studies a new codebase by reading entry points, tracing data flows, mapping module relationships, identifying patterns, and building a mental model. Produces a codebase-learner.agentcodebase-learner.mddata-engineerData processing specialist for Pandas/NumPy workflows, ETL pipelines, data validation, performance optimization, database migrations, streaming ingestion, data versioning, and big data integration. Budata-engineer.agentdata-engineer.mddatabase-reviewerPostgreSQL/Supabase database specialist for schema design, query optimization, indexing strategies, Alembic migration safety, SQLAlchemy ORM/Core patterns, connection pooling, and migration review. Endatabase-reviewer.agentdatabase-reviewer.mddependency-debt-analyzerAnalyzes dependency debt — outdated packages, known CVEs, abandoned libraries, license conflicts, excessive transitive dependencies, and version drift. Produces a structured dependency health report.dependency-debt-analyzer.agentdependency-debt-analyzer.mddoc-updaterDocumentation and codemap sync specialist. Handles OpenAPI spec regeneration, README/codemap updates, docstring refresh, architecture decision record maintenance, and example synchronization. Lightweidoc-updater.agentdoc-updater.mddocs-lookupDocumentation lookup specialist using Context7 MCP for fast, authoritative reference retrieval. Handles framework docs, library API reference, configuration guides, and best-practice lookups. Lightweidocs-lookup.agentdocs-lookup.mddocumentation-debt-analyzerAnalyzes documentation debt — stale docs, missing READMEs, undocumented public APIs, outdated comments, missing ADRs, and knowledge gaps. Produces a structured documentation debt report.documentation-debt-analyzer.agentdocumentation-debt-analyzer.mde2e-runnerEnd-to-end Playwright testing specialist for FastAPI + Next.js + Chrome extension (WXT) full-stack flows. Handles test authoring, flaky test diagnosis, CI integration, test stability, and cross-surface2e-runner.agente2e-runner.mdgit-agent-coordinatorOrchestrates multi-agent Git workflow from issue to PR. Spawns parallel research, architecture, and security agents; coordinates AI-Judge validation; creates implementation plans on feature branches; git-agent-coordinator.agentgit-agent-coordinator.mdinfra-as-code-specialistTerraform, AWS CDK, and Kubernetes specialist who owns infrastructure manifests, drift detection, GitOps with ArgoCD, policy-as-code, secrets management, infrastructure testing, and delivery pipeline infra-as-code-specialist.agentinfra-as-code-specialist.mdml-engineerML/AI/LLMOps/MLOps specialist for model training, evaluation, RAG pipelines, vector DBs, experiment tracking, hyperparameter tuning, deployment, and PyTorch build/CUDA toolchain triage. Owns the full ml-engineer.agentml-engineer.mdmulti-agent-project-managerMulti-workflow project manager agent. Continuously monitors all active multi-agent git workflows, manages agent allocation across 30+ specialists, prioritizes features from docs/features/ and Git issumulti-agent-project-manager.agentmulti-agent-project-manager.md