自然語言處理

Swift 6.0 enterprise development with async/await, SwiftUI, Combine, and Swift Concurrency. Advanced patterns for iOS, macOS, server-side Swift, and enterprise mobile applications with Context7 MCP integration.

Extension to kobo-translation skill specifically for translating video subtitles and transcripts in SRT format. Adds subtitle-specific guidelines for character limits, spoken language patterns, chunked translation context management, and maintaining readability on screen. Use this skill when translating SRT subtitle files for KoboToolbox tutorials, webinars, or educational videos.

Elasticsearch and OpenSearch skill for full-text search, aggregations, and index management. Activated when users need to search, query, or analyze data in Elasticsearch/OpenSearch clusters.

Filters PlatformIO output to reduce token usage in agentic tools. Returns success confirmation on successful builds and only error lines on failures instead of full logs. Use when agentic tools need to run PlatformIO commands to reduce context token consumption.

Use when user wants to ask Codex a question, get a second opinion, or consult with gpt-5.2. Triggers on phrases like "ask codex", "codex opinion", "what does codex think".

Sync brainstorming session discoveries to project memory (CONTEXT.md, memories)

PostgreSQL 16 database documentation. Use for SQL queries, indexes, transactions, JSON operations, full-text search, and performance tuning.

Master philosophy of language - meaning, reference, truth, speech acts. Use for: semantics, pragmatics, meaning theory, reference. Triggers: 'meaning', 'reference', 'Frege', 'sense', 'Kripke', 'speech act', 'semantics', 'pragmatics', 'truth conditions', 'propositions', 'names', 'descriptions', 'rigid designator', 'natural kind', 'context', 'indexical'.

Context-driven development methodology. Understands projects set up with Conductor (via Gemini CLI or Claude Code). Use when working with conductor/ directories, tracks, specs, plans, or when user mentions context-driven development.

Apply pattern-based bulk code transformations across multiple files. Replace console.log with logger, update imports, enforce code style, modernize syntax (var to const). Activated when users request "change all X to Y", bulk replacements, codebase-wide refactoring, or text pattern changes. Use for repetitive text-based modifications.

Create high-performing YouTube titles and thumbnail text that maximize CTR and virality while maintaining authenticity. Use when analyzing video transcripts to generate title and thumbnail suggestions, optimizing existing titles/thumbnails, or when users request help with YouTube content strategy for click-through rate optimization.

Build and deploy self-hosted Text-to-Speech API using MeloTTS from Hugging Face and create a LiveKit plugin for voice agents. Use this skill when building TTS systems, LiveKit voice agents, or self-hosted speech synthesis solutions.

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services. Use when building MCP servers to integrate APIs, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

Generate and edit images using the Gemini API (Nano Banana). Use this skill when creating images from text prompts, editing existing images, applying style transfers, generating logos with text, creating stickers, product mockups, or any image generation/manipulation task. Supports text-to-image, image editing, multi-turn refinement, and composition from multiple reference images.

Recognize when user wants a second opinion from another AI (Codex, Gemini, GLM). Asks for confirmation before calling external APIs. Triggers on phrases like "second opinion", "what would GPT say", "ask Gemini", "compare with Codex".

Assists with writing and editing general prose using Strunk & White's Elements of Style and Wikipedia's AI writing detection guide. Use when composing articles, blog posts, or editing text for clarity and authenticity.

Automatically search and retrieve relevant project memory from Mem0 when working on tasks that might benefit from past decisions, patterns, or learnings. Use this skill when starting new features, fixing bugs, or making architectural decisions to leverage historical context.

Corrects English text by removing AI-generated patterns (excessive em dashes, formal phrases), fixing speech-to-text errors (homophones, punctuation, capitalization), and improving natural flow while preserving the author's original voice and intent. Use when user asks to correct, fix, naturalize, or de-AI-ify text from any source.

[REPLACE] Apply domain-specific logic based on detected context. Use when [REPLACE with specific triggers].

Maintain AGENTS.md files across a repo: find and review every AGENTS.md, apply hygiene rules (fix outdated or incorrect guidance, reduce verbosity, split large sections into docs and link them), add newly learned context from the current conversation, and provide a concise enumerated update summary. Use when asked to update AGENTS files, run/sweep the agents update skill, or when AGENTS.md maintenance is required across a repository or subfolder.