wibandwob-dos

Arrange open WibWob-DOS windows into named layout presets: golden, magazine, cinema, triptych, diagonal, spotlight, asymmetric. Use when: "arrange windows", "lay out the desktop", "golden ratio layout", "magazine layout", "cinema mode", "triptych", "diagonal cascade", "spotlight hero window", "asymmetric layout", "Swiss grid", "arrange like a magazine", "make it look nice", "layout preset". Ports the arrange.py presets from wibandwob-dos-tvision. Each preset is a pure math function over window positions — no blessed, no SDK, just batch move/resize.

Convert any URL to clean Markdown instantly. Uses Defuddle for content extraction, falls back to pi-web-browse automatically on 403/bot-block. Use when: 'fetch this URL', 'convert to markdown', 'read this page', 'get article content', 'extract this URL'.

TERMINAL EMULATOR LAYER — control Ghostty itself, below WibWob-DOS. Uses AppleScript to act as a human at the keyboard/mouse: click blessed menu items, send keystrokes, inject shell commands, take PNG screenshots, ghost-click TUI cells. Use when: clicking a menu, sending a keystroke, restarting the TUI, smoke-testing the UI, simulating user input. NOT for: opening microapps, reading window content, or anything the HTTP API can do — use wibwobdos-control for that. Typical sequence: ghostty-control to act → wibwobdos-control to verify. macOS only. Requires Ghostty >= 1.3.0. Scripts: calibrate.sh, click-cell.sh, click-text.sh, menu-click.sh, send-to-terminal.sh, wait-for.sh, restart-wibwob.sh, ghostty-windows.sh.

Surface open ideas and contribute friction back. Run scripts/devlog-open.sh to see unresolved pains and ideas from this week's reflection. Use at the start of any session, or when you notice something worth recording. Triggers on: "devlog", "briefing", "what's been happening", "catch me up", "open ideas", "what should I work on".

APPLICATION LAYER — control what runs inside Ghostty, not Ghostty itself. Talks to the WibWob-DOS HTTP API (port 8099) to open/move/read windows, check desktop state, send messages, screenshot, share to Discord, manage shaders. Use when: opening a microapp, reading a window's content, checking desktop layout, sending a command, managing shaders. NOT for: clicking menus, sending keystrokes, or interacting with the terminal emulator itself — use ghostty-control for that.

Scaffold, register, and dev-loop a new WibWob-DOS microapp end-to-end. Use when: "create a microapp", "new microapp", "scaffold a microapp", "add a microapp", "build a new app for wibwob", "make a wibwob app".

Reflect on implementation friction after sessions. Write for yourself and future agents. Focus on pain → why → fix. Skip git-log style entries.

Share WibWob-DOS TUI to a Discord channel. Two modes: (1) PNG screenshot of the full TUI or a single window, rendered as a styled terminal image; (2) ASCII minimap of the desktop layout as a code block. Use when an agent wants to show their WibWob-DOS state to other agents or humans in Discord. Requires DISCORD_WEBHOOK_URL env var.

Read this skill before making git commits. Enforces WibWob-DOS conventional commit format.

Add a new command to WibWob-DOS — walks the 4-file wiring path so you don't miss a step. Use when: "add a command", "new command", "wire a command", "scaffold command".

Radically simplify documentation by applying the system-knowledge test: if a script, skill, CLI, or the running system already tells you something, don't write it down. Docs contain decisions, constraints, and conventions only. Use when: "simplify docs", "tighten specs", "agent-friendly docs", "progressive disclosure", "dedupe docs", "too many docs", "doc cleanup".

Record choreographed WibWob-DOS TUI sessions as retina GIF/MP4 with synced chiptune SFX audio. Use when recording a demo, making a show reel, capturing a choreographed sequence, exporting video from the TUI, or producing any polished visual artifact of the running desktop. Also use when the user asks to "record this", "make a video", "export a GIF", "capture a reel", or mentions "cinema", "wibwob-record", or "make an MP4".

Autopoietic documentation agent. Runs in a fresh context window with delta-compressed system context. Reads AGENTS.md (system describes itself), makes one targeted improvement, validates via doc-health.sh (system measures itself). Output must be valid system artifacts — the agent maintains the system that described it. Use when: "run the doc agent", "improve doc health", "self-maintain docs", "doc-agent", "autopoietic agent".

Monthly repo health check — branch census, worktree audit, stale docs, planning drift, GH issue triage. Run when branches pile up, worktrees sprawl, or planning docs drift from reality. Triggers on: "repo hygiene", "branch cleanup", "jubilee", "audit branches", "stale docs", "planning drift".

Compose and execute timed music video shows inside WibWob-DOS TUI. Declarative timeline files sync visual cues (window layout, theme changes, primer art, figlet typography) to audio playback. Use when: creating a music video, VJ show, timed visual performance, or any task that needs coordinated window choreography synced to audio. Triggers on: "music video", "VJ show", "visual performance", "timeline", "timed show", "sync to music".

Autonomously optimize any Claude Code skill by running it repeatedly, scoring outputs against binary evals, mutating the prompt, and keeping improvements. Based on Karpathy's autoresearch methodology. Use when: optimize this skill, improve this skill, run autoresearch on, make this skill better, self-improve skill, benchmark skill, eval my skill, run evals on. Outputs: an improved SKILL.md, a results log, and a changelog of every mutation tried.

Update .planning docs after completing work — tick story checkboxes, advance story/feature/epic status, run planning:sync, commit. Use after finishing any story, feature, or epic. Triggers on: task done, story complete, epic closed, closing out work, updating planning docs.

Run end-to-end hosting-agnostic smoke validation for WibWob-DOS across Docker/VPS/Fly.io/npm-global style deployments using adapter runners + shared binary gates (CLI/API/screenshot/persistence/external microapps), then produce an agent-readable report with pass/fail outcomes and remediation notes.

Maintain a CHANGELOG.md that follows the Keep a Changelog 1.1.0 convention. Use when: adding a changelog entry, releasing a version, creating a CHANGELOG.md from scratch, updating the changelog after a commit, bumping a version, writing release notes, or when the user says "update the changelog", "add to changelog", "what changed", "log this release", "keep a changelog", or "write release notes".

Create well-structured Agent Skills that follow the agentskills.io specification and best practices. Use when building a new skill, restructuring an existing one, writing a SKILL.md, or when the user says "make a skill", "create a skill", "scaffold a skill", "write a SKILL.md", or asks how to structure skill directories, frontmatter, descriptions, scripts, or progressive disclosure.

Library of 500+ ASCII art pieces by Joan G. Stark (jgs), legendary ASCII artist 1996-2001. Use to find period-authentic ASCII art for primers, decoration, and composition. Browse the examples directory and open as primer windows. Art covers animals, birds, holidays, food, people, borders and more.

Export recent git commits grouped by branch into a plain text report. Use when the user asks for "git oneliners", "last 24h commits by branch", "repo summary","branch commit summary", or wants output like "# branch" then "- sha — MM-DD HH:MM — message".

Audit local Pi skill, extension, and agent usage from session logs for the current repo. Flags stale items not used in N days and outputs an agent-friendly Markdown/JSON report. Use when asked what is stale, unused for 2 weeks, or to rank under-used skills/extensions/agents.

Mine pi agent session logs (~/.pi/agent/sessions/) to find which subsystems cause the most agent confusion, regressions, and repeated human re-explanation. Three-step pipeline: extract JSONL to searchable markdown, regex-scan for quantitative trouble ranking, then LLM deep-analysis of worst sessions for semantic patterns regex can't catch. Use when: auditing agent experience, planning which subsystems need specs first, building trigger tables, investigating why agents keep getting something wrong, or feeding a codified context infrastructure (E001 pattern).

Manage the signls generative MIDI sequencer. Install, launch in tmux, stop, check status, and inspect bank files. Use when composing generative MIDI sequences, launching the sequencer, managing bank files, or any task involving signls.

Composable audio synthesis toolkit for chiptune music, covers, and scoring. Two workflows: BRICKS (original compositions from numpy primitives) and COVER (arrange well-known melodies as chiptune). Use when composing music, generating stingers, mixing voiceover with score, covering songs in 8-bit, or any task requiring programmatic audio synthesis.

Delegate tasks to OpenAI Codex CLI as a subagent. Two agents — codex-standard (gpt-5.3-codex, general-purpose) and codex-heavy (gpt-5.4, complex tasks). USAGE RESTRICTION — Codex budget is limited this month. Do NOT use proactively. Only use when the human explicitly requests Codex. Ask permission before delegating.

Convert images to plain-text ASCII art primers for the TUI. Covers sourcing images, tuning conversion parameters, and opening the result as a primer window in the running app.

Search and extract content from pi agent session logs (~/.pi/agent/sessions/). Use to recover lost work (files written by agents), find what was discussed, replay commands the agent ran, or audit what a wibwob-agent session did. Triggers on: "what did the agent write", "recover lost module", "find that thing the agent made", "search session logs", "what happened in that session", "did the agent commit that", "recover from session".

Composable audio synthesis toolkit for building chiptune music from numpy primitives. Three layers: oscillators/effects/theory (raw materials), patterns (compositional gestures), pipeline (pydub bridge, mixing, splicing, voice, export). Use when composing original music for video reels, generating stingers/interstitials, mixing voiceover with score, or any task requiring programmatic audio synthesis. Triggers on: music composition, chiptune generation, audio scoring, sound design, stinger creation, reel audio mixing, chiptune studio, studio, open the studio, audio synthesis, compose audio.

Fetch the transcript of any YouTube video as plain text. Use when user shares a YouTube URL and wants it summarised, analysed, quoted from, or processed in any way. Triggers on: "transcribe this", "get the transcript", "summarise this video", "what does this video say", any YouTube URL shared without further context.

Create typographic video sequences using figlet text animations in WibWob-DOS. Use when composing figlet-based title cards, animated text sequences, or typographic VJ elements synced to music.

A director-lens inspired by Michel Gondry’s public body of work and process: - handmade, practical, “visible craft” effects - playful logic, loops, repetition, and rule-based worlds - music mapped to choreography, objects, or environment “scores” - simple ideas executed with obsessive structure

Smoke test a VJ timeline end-to-end with real macOS screencapture PNGs at every cue step, plus tmux text dumps and state JSON. Capture-then-review pattern — all evidence collected first, comparison done after. Use when testing a new timeline, verifying the pipeline after code changes, or debugging cue behaviour. Triggers on: "smoke test timeline", "test the timeline", "capture each step", "screenshot each cue".

Search and extract ASCII art, themes, primers, dialogue, and metadata from the Wib&Wob Backrooms raw archive (662 session logs, 57MB). Use when looking for specific art, finding sessions by primer/theme/model, extracting ASCII blocks, or exploring the creative corpus.

Convert well-known melodies, themes, and scores into chiptune arrangements using chiptune-studio. Takes a song reference (title, artist, or melody description), researches key/tempo/notes, makes creative arrangement decisions per scene or section, and outputs a composed WAV via chiptune-studio synthesis. Use when: user asks to "make a chiptune version of [song]", "cover [track] in 8-bit", "arrange [theme] for chiptune", "do a chiptune [song name]", or references any well-known melody that should be rendered through the chiptune-studio toolkit. Also triggers on: "soundtrack this with [song]", "score this video with a chiptune [reference]".

Search markdown knowledge bases, notes, and documentation using QMD. Use when users ask to search notes, find documents, or look up information.

Three-pass review of planning epics, features, stories, spikes, and refactor trackers. Checks execution coherence (status fields vs reality, AC-to-test traceability, lifecycle sync), scope hygiene (stories that are actually three stories, ACs that are untestable, scope creep), and canon compliance (naming rules, checkbox format, commit conventions, closeout pass rules from AGENTS.md and .planning/README.md). Fixes issues directly. Use when the user says "simplify planning", "tidy the epics", "audit the briefs", or before closing out a feature or epic.

Three-pass code review and cleanup after a batch of changes. Runs reuse analysis (find existing utilities that replace new code), quality review (redundant state, copy-paste, leaky abstractions, stringly-typed code), and efficiency review (unnecessary work, missed concurrency, hot-path bloat, memory leaks). Fixes issues directly. Use after landing a feature slice, before a PR, or when the user says "simplify", "clean up", or "review".