Pydantic plugins

Build scalable production Python backends and APIs with Django 5.x async views, FastAPI microservices, Celery tasks, SQLAlchemy/Pydantic data handling, pytest testing strategies, and architecture optimizations using uv/ruff for modern 3.12+ codebases.

Run institutional-grade stock analysis on A-share, HK, and US equities with 22 financial dimensions, simulated investor panels, DCF/LBO valuation models, and pig-butchering scam detection, all output as Bloomberg-style HTML reports.

Build sophisticated AI agents using LangChain, LangGraph, and Deep Agents: configure memory and filesystem backends, orchestrate subagents for task delegation, add human-in-the-loop approvals via middleware, construct RAG pipelines, and manage persistent state graphs with checkpointers.

Profile Python performance bottlenecks with cProfile/py-spy, analyze pytest test suites for quality/coverage, check async code for issues/patterns, lint/fix with ruff, optimize algorithms/memory, generate unit/integration tests, and package/publish projects using uv/pyproject.toml.

Scaffold new MCP servers in TypeScript or Python with configurable transports, Zod/Pydantic validation, error handling, and example tools/prompts, then add tools to existing servers via guided steps for schemas, handlers, registration, testing, and docs.

Orchestrate end-to-end DSPy 3.2.x workflows to build, evaluate, optimize, and deploy production-grade coding agents, using typed signatures, modules, rich-feedback harnesses, GEPA optimization, RLM for long-context reasoning over codebases or docs, and full project automation from scratch.

Generate interactive marimo notebooks from arXiv research papers, migrate Jupyter and Streamlit apps to marimo format, build reactive anywidget UI components with JavaScript, add GitHub sharing badges, prepare notebooks for batch scheduling with Pydantic params, and verify WASM compatibility for web deployment.

Establishes opinionated Python 3.11+ engineering standards with SOLID principles, strict typing, pytest testing, ruff linting; automates TDD workflows, routes to specialists for CLI apps (Typer/Rich/Textual), web APIs (FastAPI/Flask/Django), data pipelines, packaging, code reviews, and PyPI CI/CD deployment.

Build modern Python web apps with Django and FastAPI, define SQLAlchemy models and Alembic migrations, write pytest tests, debug errors, and review code quality using specialized agents that orchestrate database tasks and full codebase reviews.

Audit codebases for 12-Factor App and Agent compliance, providing ratings, evidence from files, gaps, and recommendations. Extract decisions from conversations to generate MADR ADRs. Brainstorm raw ideas into detailed project specs. Compare git repos via LLM-as-judge scoring on functionality, security, tests, and overengineering.

Develop production-grade Python backends with advanced type hints, mypy static analysis, async patterns like gather and create_task, and data modeling via dataclasses, attrs, or Pydantic for validation and serialization. Use Pyright LSP for real-time diagnostics, navigation, and refactoring in .py files.

Autonomously draft complete USPTO, EPO, and PCT patent applications from invention disclosures: search 100M+ patents via BigQuery for prior art and patentability, analyze claims/specifications for MPEP/35 USC 112/EPC Art 84 compliance, generate Graphviz diagrams, validate formalities, and output markdown+SVG filing packages. Self-contained, no MCP server needed.

Engineer robust ETL pipelines: clean messy CSVs/Parquet, infer schemas, profile datasets, detect anomalies, validate quality with Pydantic/Pandera/Great Expectations, implement incremental patterns, generate dbt models/SQL migrations/tests, and orchestrate autonomous backfills/pipeline testing via agents and CLI commands.

Dispatch AI agents to forecast, score, classify, research, deduplicate, merge, and filter rows in Python pandas DataFrames at scale using natural language instructions. Access remote MCP endpoints for search tools and Everyrow shared databases for queries, schema inspection, and manipulation via API keys.

Author Claude Code artifacts like skills, rules, hooks, commands, CLAUDE.md files, and scripts using guided best practices for prompt structuring, structured JSON outputs, multi-agent orchestration, automation scripts, context optimization, quality evals, and tool interfaces.

Build production-grade Python applications using FastAPI for REST APIs, SQLAlchemy for ORM, Temporal for durable workflows, functional core/imperative shell architecture, stub-driven TDD with pytest, monorepo setups via uv/mise/ruff, and git workflows for stacked commits/reviews.

Develop, configure, deploy, and debug serverless Python apps on Modal.com: set up GPU-accelerated functions (T4-H200) with autoscaling, create FastAPI/ASGI endpoints and cron schedules, manage sandboxes/volumes/secrets, integrate GitHub Actions CI/CD, and troubleshoot failures for AI/ML workloads.

Scaffold production-ready REST and GraphQL APIs using FastAPI or Django, with async patterns, database integration, middleware, authentication, and scalable microservices architectures. Delegate to expert agents for backend design, performance optimization, WebSockets, gRPC, and event-driven systems.

Build production-ready AI agents and multi-agent teams with the Agno framework by creating workflows and MCP integrations, adding tools, memory, debugging, and structured outputs, then deploy them via AgentOS using FastAPI.

Build, design, and troubleshoot Python CLI applications with Typer using type-hint driven workflows for command structures, input validation, pytest testing, and distribution packaging.

Summon Python specialists to scaffold production Django/FastAPI projects with uv/Docker/PostgreSQL, enforce Mypy/PEP8/security reviews, audit codebases for multi-agent parallelization, implement Celery tasks/WebSockets, and generate pytest strategies—all using 2025 patterns and official docs.

Audit repositories and refactor code with zero-tolerance pedantry: enforce precise naming, strict casing laws, structural symmetry, import discipline, declaration order, no magic values or dead code, and consistent patterns in Python, TypeScript, Go, and JavaScript projects. Generate detailed reports, compliance verdicts, and fix suggestions with pedantry scores.

Scaffold opinionated Django projects with one-file-per-model organization, Ninja APIs via domain-grouped routers and Pydantic schemas, Unfold admin with HTMX and Tailwind, pytest tests using factory_boy, Dynaconf config, uv deps, and Docker setup. Delegate code reviews to the agent to enforce patterns after changes.

Secure Python-based MCP servers and multi-agent pipelines by implementing OAuth 2.0 flows with PKCE and providers like Google/GitHub, enforcing 5-layer defenses against prompt/SQL injections and unauthorized access via RBAC, and achieving GDPR/CCPA compliance with consent management, data minimization patterns, and regulatory checklists.