TOOLS BY MING
Small tools.
Useful work.
Products you can try and skills you can use with your AI assistant. Each starts here on SideCourt.
Web apps open in your browser. Skills run in a compatible AI assistant; the code remains on GitHub. Legacy skills are retained for reference.
20 tools and skillsRedBot
Choose one useful saved source, read the relevant passage and take a concrete next action with a mobile web beta or an AI-assistant skill.
See how it works ↗Web app · AvailableDayCup
Get a coffee starting recipe for your equipment and beans, then keep your brew notes.
See how it works ↗Desktop app · AvailableCleanPause
Pause device input while you clean your keyboard and trackpad.
See how it works ↗Web app · Account and key requiredEarBrief
Turn selected text and reports into audio you can listen to away from the screen.
See how it works ↗Skill · AvailableDesign Agent
Turn visual references into focused interface changes, actionable implementation briefs and verified builds within the current project scope.
See how it works ↗Skill · LegacyProject Rebuild Advisor
Project rebuild advisor that inspects a confused or inherited project and tells you whether to recover, clarify, repair, rebuild, continue, or stop.
See how it works ↗Skill · AvailableProject Lessons Log
Project lessons log that records costly incidents, distills reusable lessons, and prevents expired decisions from contaminating the next project.
See how it works ↗Skill · AvailableIteration Cleanup & Handoff
Iteration cleanup and handoff that preserves current decisions, removes superseded work safely, verifies a cold takeover, and prepares the next milestone.
See how it works ↗Skill · LegacyProduct System Map
Product system map for ideas and codebases: connect user journeys, capabilities, responsibilities, missing paths, and buildable delivery slices at the scope you need.
See how it works ↗Skill · AvailablePlaylist Builder
Playlist builder that turns a mood, scene, or seed songs into a verified playlist you can import to Apple Music, Spotify, YouTube Music, or other services.
See how it works ↗Skill · AvailableFirst-Time User Tester
Walk through a product as a first-time user and produce an evidence-led HTML report with task outcomes, prioritized fixes and concrete retests.
See how it works ↗Skill · AvailablePrompt Cleaner
Prompt cleaner for dictated or mixed-language notes. Produce a reusable AI prompt while preserving corrections, examples, scope, and calendar-week boundaries.
See how it works ↗Skill · AvailableGitHub Skill Publisher
Review a skill from the user’s perspective, learn from comparable skills, test improvements, and publish a verified GitHub product and install package.
See how it works ↗Skill · AvailableAI Instruction Stress Test
AI instruction stress test that simulates users and environments to expose trigger misses, skill collisions, stale assumptions, and silent behavior failures before release.
See how it works ↗Skill · LegacyDecision Thinking Partner
Decision thinking partner that clarifies ambiguous choices, keeps real alternatives, recommends only when evidence supports it, and on request grills a build plan into a locked scope.
See how it works ↗Skill · AvailableTaskDock
Keep agent work ready to continue: portable task identity, safe file organization, bounded resume and a readable private handoff page.
See how it works ↗Skill · AvailableStock Portfolio Brief
Turn stock holdings and fresh filings into a source-backed portfolio decision, with index comparisons, thesis checks and explicit data gaps.
See how it works ↗Skill · AvailablePodcast Video Maker
Turn video, text or notes into a directed podcast conversation, with budgeted voice generation and lower-cost source-video lip sync.
See how it works ↗Skill · AvailableTask Execution Coordinator
Translate human intent into scoped work for existing owners, select relevant capabilities and evidence, and verify outcomes without adding a second task queue.
See how it works ↗Skill · AvailableWork Memory
Recover current project decisions with sourced recall, outcome-tested lessons and read-only readiness reports that expose stale evidence and missing checkpoints.
See how it works ↗