前端開發

High-performance vector similarity search engine for RAG and semantic search. Use when building production RAG systems requiring fast nearest neighbor search, hybrid search with filtering, or scalable vector storage with Rust-powered performance.

GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.

Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.

Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.

LLM observability platform for tracing, evaluation, and monitoring. Use when debugging LLM applications, evaluating model outputs against datasets, monitoring production systems, or building systematic testing pipelines for AI applications.

Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework

Build complex AI systems with declarative programming, optimize prompts automatically, create modular RAG systems and agents with DSPy - Stanford NLP's framework for systematic LM programming

Handles all Golang testing tasks including running tests, writing new tests, and fixing test failures. Follows MCPSpy testing conventions with require for critical assertions and assert for non-critical ones.

Static analysis of UEFI/BIOS firmware dumps using Intel's chipsec framework. Decode firmware structure, detect known malware and rootkits (LoJax, ThinkPwn, HackingTeam, MosaicRegressor), generate EFI executable inventories with hashes, extract NVRAM variables, and parse SPI flash descriptors. Use when analyzing firmware .bin/.rom/.fd/.cap files offline without requiring hardware access.

Turns product or tech specs into concrete Notion tasks that Claude code can implement. Breaks down spec pages into detailed implementation plans with clear tasks, acceptance criteria, and progress tracking to guide development from requirements to completion.

Performs comprehensive code reviews for Bitwarden iOS projects, verifying architecture compliance, style guidelines, compilation safety, test coverage, and security requirements. Use when reviewing pull requests, checking commits, analyzing code changes, verifying Bitwarden coding standards, evaluating unidirectional data flow pattern, checking services container dependency injection usage, reviewing security implementations, or assessing test coverage. Automatically invoked by CI pipeline or manually for interactive code reviews.

DOM Testing Library patterns for behavior-driven UI testing. Framework-agnostic patterns for testing user interfaces. Use when testing any front-end application.

React Testing Library patterns for testing React components, hooks, and context. Use when testing React applications.

Batch-generate images via OpenAI Images API. Random prompt sampler + `index.html` gallery.

Transform vague feature ideas into clear, testable requirements using EARS format. Capture user stories, define acceptance criteria, identify edge cases, and validate completeness before moving to design.

Systematic three-phase approach to feature development using Requirements, Design, and Tasks phases. Transforms vague feature ideas into well-defined, implementable solutions that reduce ambiguity, improve quality, and enable effective AI collaboration.

Transform approved requirements into comprehensive technical designs. Define system architecture, component interactions, data models, and interfaces to create a blueprint for implementation.

Multi-agent autonomous startup system for Claude Code. Triggers on "Loki Mode". Orchestrates 100+ specialized agents across engineering, QA, DevOps, security, data/ML, business operations, marketing, HR, and customer success. Takes PRD to fully deployed, revenue-generating product with zero human intervention. Features Task tool for subagent dispatch, parallel code review with 3 specialized reviewers, severity-based issue triage, distributed task queue with dead letter handling, automatic deployment to cloud providers, A/B testing, customer feedback loops, incident response, circuit breakers, and self-healing. Handles rate limits via distributed state checkpoints and auto-resume with exponential backoff. Requires --dangerously-skip-permissions flag.

JavaScript and TypeScript development with ES6+, Node.js, React, and modern web frameworks. Use for frontend, backend, or full-stack JavaScript/TypeScript projects.

Writing effective code documentation - API docs, README files, inline comments, and technical guides. Use for documenting codebases, APIs, or writing developer guides.