@daymade

cloudflare-troubleshooting provides systematic api-driven investigation for common errors (redirect loops, dns issues, ssl certificate failures, origin errors). emphasis on evidence gathering before diagnosis, with curl examples and decision logic for each category.

macos-cleaner provides systematic disk analysis via mole with safety-first deletion policies. it prioritizes user confirmation, impact analysis, and preservation of valuable caches over aggressive space reclamation.

fact-checker identifies and verifies factual claims in documents against authoritative sources, then proposes corrections with explicit user approval before applying edits. covers technical specs, version numbers, and statistical data.

claude-skills-troubleshooting covers plugin installation and activation diagnostics, addressing the common mismatch between installed and enabled plugin states with concrete CLI commands and manual fixes.

capture-screen provides a three-step macOS workflow to enumerate windows via Swift CoreGraphics, control application state with AppleScript, and capture PNG/JPEG output using screencapture. best for excel automation but generalizes to any on-screen window.

video-comparer generates interactive html reports by extracting metadata and quality metrics (psnr, ssim) from paired videos, supporting three viewing modes and frame-by-frame visual analysis with configurable sampling intervals.

ui-designer extracts design systems from reference images and generates implementation-ready prompts by analyzing visual patterns, generating documentation, and composing tailored ui implementation instructions.

youtube-downloader handles video and audio extraction from youtube and hls streams via yt-dlp and ffmpeg, with fallbacks for authentication, quality negotiation, and common failure modes like bot detection and format unavailability.

i18n-expert covers framework selection, locale file setup, string replacement, and audit validation for react/next/vue codebases targeting en-us and zh-cn with configurable translation quality strategies.

promptfoo-evaluation covers configuring and running llm evaluations with promptfoo, including prompt formats, test case setup, custom assertions, llm-as-judge grading, and relay api integration with explicit gotchas around file path resolution and concurrency settings.

claude-code-history-files-finder recovers deleted or modified code from claude code session history stored locally, supporting keyword search, deduplication, and batch content extraction with verification workflows.

financial-data-collector retrieves standardized financial metrics for us public companies via yfinance. output is json with mandatory source attribution and explicit null handling for missing fields, designed for downstream consumption.

prompt-optimizer converts unstructured requirements into testable specifications by applying EARS syntax patterns and domain-specific frameworks, then generating structured prompts via a four-layer enhancement pipeline.

product-analysis spawns parallel claude code and codex cli agents to audit ux, api, and architecture across multiple dimensions, synthesizing findings into a prioritized report with cross-model validation.

repomix-safe-mixer detects and removes hardcoded credentials before packaging codebases with repomix, reducing accidental exposure risk during distribution and sharing workflows.

cli-demo-generator produces animated GIF demos from terminal recordings, offering four methods ranging from fully automated command execution to manual control. incomplete specification leaves the other three approaches unexplained.

excel-automation covers file creation with openpyxl, complex xlsm parsing via zipfile and xml, and macos applescript control. targets investment banking workflows with formatting and corruption recovery.

qa-expert outlines triggers for test infrastructure, case design, and security testing workflows, but omits procedural depth and failure handling needed for reliable automation.

github-ops covers pull requests, issues, workflows, and api interactions but lacks concrete procedural guidance and failure mode handling. provides activation context without actionable implementation.

transcript-fixer applies two-phase correction: dictionary rules first, then ai-powered analysis. builds a personalized corrections database that accumulates fixes to improve accuracy incrementally.

skill-reviewer validates claude code skills against best practices. incomplete specification, vague modes, and a setup section that cuts off mid-procedure. lacks actionable review criteria and clear success metrics.

llm-icon-finder locates and constructs cdn urls for ai model and provider brand icons from the lobe-icons github library, supporting svg, png, and webp formats across light and dark theme variants.

developing-ios-apps documents xcodegen and swift package manager workflows with heavy emphasis on signing and notarization failure modes, but lacks procedural depth and actionable step sequences for core build tasks.

twitter-reader fetches x posts and articles with media, but lacks structured decision logic, explicit failure handling, and complete procedure documentation beyond a single quick-start example.

skills-search delegates to ccpm CLI for registry operations. the file specifies execution behavior and bootstrap logic but lacks concrete search/discovery use cases and outcome signals.

deep-research describes a multi-agent architecture for synthesizing research reports but lacks concrete procedural steps, decision logic, input specifications, or output contracts.

Convert markdown files to professional PDF documents with proper Chinese font support, theme system, and visual self-check. Use whenever the user asks to…

|

Install, configure, customize, or troubleshoot the Claude Code statusline (the line above the prompt with cwd, model, and token counts). Use when the user…

|

Extracts Mermaid diagrams from markdown files and generates high-quality PNG images using bundled scripts. Activates when working with Mermaid diagrams,…

Extracts files from repomix-packed repositories, restoring original directory structures from XML/Markdown/JSON formats. Activates when users need to unmix…

Creates educational Teams channel posts for internal knowledge sharing about Claude Code features, tools, and best practices. Applies when writing posts,…

End-to-end playbook for shipping high-quality pull requests to open-source projects you don't maintain. Use whenever the user is creating, editing, or pushing…

docs-cleaner — an installable skill for AI agents, published by daymade/claude-code-skills.

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an…

Create professional slide decks from topics or documents. Generates structured content with data-driven charts, speaker notes, and complete PPTX files. Applies…