Claude Skills

Author Smithers (smithers-orchestrator) Bun/TSX JSX workflows for Claude Code: durable, resumable, observable orchestration with SQLite state (phases/steps), tool-scoped agents, optional parallelism, and workflow logs.

Run the usage query script to retrieve account usage information for GLM Coding Plan. Only use when invoked by usage-query-agent.

Diagnose and fix broken builds or failing tests with coordinated retries. Use when a build/test is broken, pre-commit hooks fail, or agents should coordinate a fix.

Semantic code analysis via LSP. Navigate code (definitions, references, implementations), search symbols, preview refactorings, and get file outlines. Use for exploring unfamiliar codebases or performing safe refactoring.

Expert prompt engineering for AI systems. Use when the user wants to write or review prompts for AI, create instructions for AI systems, build system prompts, review or improve existing prompts, optimize AI instructions, or create any form of written communication intended for AI consumption (Claude, GPT, or other LLMs).

Expert-level Kamal deployment guidance for deploying containerized applications to any server. Use this skill when users ask about Kamal, container deployment, zero-downtime deployments, deploying Rails/web apps to VPS/cloud servers, kamal setup, kamal deploy, Docker deployment without Kubernetes, or deploying to Hetzner/DigitalOcean/AWS with Kamal. Also use when users mention DHH's deployment tool, 37signals deployment, or want an alternative to Heroku/Render/Vercel with self-hosted infrastructure.

Comprehensive toolkit for generating, validating, and managing Kubernetes YAML resources. Use this skill when creating Kubernetes manifests (Deployments, Services, ConfigMaps, StatefulSets, etc.), working with Custom Resource Definitions (CRDs), or generating production-ready K8s configurations.

Comprehensive Kubernetes debugging and troubleshooting toolkit. Use this skill when diagnosing Kubernetes cluster issues, debugging failing pods, investigating network connectivity problems, analyzing resource usage, troubleshooting deployments, or performing cluster health checks.

Create annotated blog posts from technical talks with slides. Use when asked to convert a video presentation to a blog post, create written content from a talk, or annotate slides with transcript.

Fetch Kit (ConvertKit) newsletter broadcasts for writing context. Use when asked to download newsletters, get past email content for style reference, or fetch broadcasts for analysis.

Query and sync Granola meetings to Obsidian vault. Use when user mentions Granola, meeting transcripts, or wants to sync meeting notes. Reads from local cache - no API needed.

Manage tasks in Obsidian via TaskNotes plugin API. Use when user wants to create tasks, list tasks, query by status or project, update task status, delete tasks, or check what they need to do.

Download YouTube video transcripts with timestamps. Use when asked to transcribe a YouTube video, get transcript, or extract text from a video URL.

Analyze Wispr Flow voice dictation data. Stats, search, export, visualizations. Use when user says "dictation history", "word counts", "voice analytics", "how much did I dictate", "search my dictation".

Query Obsidian Bases via the Bases Query plugin (RPC). Use when you need to read structured data from Obsidian bases.

Generate chapter summaries with timestamps for YouTube videos using AI. Use when asked to create chapters, summarize video sections, or generate video outline.

Multimodal AI processing using Google Gemini. Use for analyzing PDFs, images, videos, YouTube links, and other large documents. Ideal when you need to extract information from files that require vision or multimodal understanding.

Download Zoom meeting transcripts. Use when asked to get transcripts from Zoom recordings, download Zoom meeting notes, or fetch Zoom call transcripts.