Skip to content
domesticmousePublic

About

Agent Skills

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

22 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Skills

A collection of AI agent skills by @domesticmouse, compatible with skills.sh.

Installation

Install all default skills or pick interactively:

npx skills add domesticmouse/skills

Install a specific skill:

npx skills add domesticmouse/skills --skill <skill-name>

Install an optional skill pack (not installed by default):

npx skills add domesticmouse/skills/packs/<pack-name>

Available Skills

Discovers, lists, inspects, compares, audits, exports, and safely deletes Google Antigravity session trajectories, transcripts, disk paths, tool invocations, and execution steps across all brain storage directories under ~/.gemini.

Origin: Built around a zero-setup uv Python CLI script (agy-brain-explorer.py) to make navigating local Antigravity transcript.jsonl and transcript_full.jsonl session logs fast and readable directly inside chat via GitHub Flavored Markdown tables and clickable conversation:// and file:// links.

Intended Usage: Ask the agent to list recent sessions, group or search sessions by project or prompt keyword, inspect step timelines and tool calls, export per-conversation Markdown logs (e.g., into a prompts/ folder), compare two sessions side-by-side, or clean up old sessions (moving them to the macOS Trash by default).

Guides the agent in writing, editing, or reviewing blog posts, tutorials, articles, and social media updates in Brett Morgan's authentic writing voice.

Origin: Distilled from analysis of Brett's existing articles, technical write-ups, and social posts to capture recurring linguistic signatures, Australian English spelling, and conversational cadence.

Intended Usage: Activated when drafting, editing, or reviewing blog posts, tutorials, articles, or social media updates that should sound like Brett.

Crops, trims, and adjusts image framing by removing transparent whitespace margins, trimming solid color borders, or extracting specific pixel bounding boxes.

Origin: Built around a zero-dependency Python 3 utility (crop.py) that decodes and re-encodes PNGs using only the Python standard library (with automatic fallback to Pillow for JPEG/WebP) to inspect subject bounding boxes and perform lossless crops.

Intended Usage: Activated when inspecting image dimensions and subject margins, autocropping transparent or solid-color borders with configurable padding, or cropping images to exact pixel coordinates.

Plans, walks through, captures, and verifies iOS Simulator screenshots using xcrun simctl io booted screenshot with full absolute paths.

Origin: Created to standardize staging clean, zero-margin device-framebuffer screenshots (artifacts/screenshots/<project_name>/) by discovering an app's actual views/screens and generating copy-pasteable simctl walkthroughs that work from any active terminal directory.

Intended Usage: Activated when the user asks where to store mobile app screenshots, which screens to capture for an iOS or Flutter app run, wants a step-by-step xcrun simctl screenshot walkthrough, or asks to visually and dimensionally verify captured simulator PNGs alongside /crop-image and /slidev-deck.

Scaffolds, authors, styles, previews, and deploys dark-theme Slidev presentation decks (slides/) with modular act files, custom Vue layouts, and embedded asciinema-player terminal recordings.

Origin: Extracted from building technical presentation decks combining modular Slidev acts, high-contrast Tokyo Night / Dracula terminal themes, SSR-safe AsciinemaPlayer.vue wrappers with adaptive segment speed compilation, opt-in development chapter stepper bars (showChapterBar), BASE_URL asset resolution, custom zoom-in and visual callout layouts, background dev-server management (slides-server.sh), and GitHub Pages CI deployment.

Intended Usage: Activated when creating a new Slidev deck, authoring or refactoring slides and conversational presenter notes, embedding .cast terminal recordings or visual callouts, running the local Slidev dev server, or configuring GitHub Pages deployment.

Recommended Companion Pack: Pair with packs/asciinema (npx skills add domesticmouse/skills/packs/asciinema) when recording, sanitizing, trimming, or chapter-marking .cast terminal sessions for Slidev.

An interactive, stateful code-teaching workflow that builds self-contained HTML lessons (compiled from Markdown via build.py), printable reference sheets, learning records, and runnable per-lesson companion projects in apps/ verified with automated tests and linters.

Origin: Derived from Matt Pocock's original teach skill and refined across multiple hands-on technical courses to incorporate Tufte-inspired dark/print typography, Prism.js keep-markup fill-in-the-blank exercises, Mermaid.js diagrams, equal-length retrieval quizzes, block-by-block code walkthroughs, reference schema templates (LESSON-FORMAT.md, MISSION-FORMAT.md, RESOURCES-FORMAT.md, LEARNING-RECORD-FORMAT.md, GLOSSARY-FORMAT.md), a companion listing synchronizer (sync_listing.py), and automated verification gates.

Intended Usage: Invoke with /teach-code (or ask the agent to teach a programming language, framework, or engineering tool) to establish a MISSION.md, curate RESOURCES.md, and progressively build interactive lessons backed by verified, runnable code.

Skill Packs

Skill packs live under packs/ so they are excluded from default npx skills add domesticmouse/skills installs, while remaining installable as a group via subpath.

packs/asciinema — Terminal Recording & Interactive Cast Player Pack

Skills and zero-dependency Python tooling for recording, sanitizing, trimming, chapter-marking, and test-driving interactive asciinema terminal sessions:

npx skills add domesticmouse/skills/packs/asciinema

packs/egui — Engine-First Rust & egui Application Pack

A cohesive set of 5 skills for architecting, building, driving, and verifying multi-crate Rust + egui/eframe/wgpu applications across native desktop, headless CLI/MCP, and WebAssembly targets:

npx skills add domesticmouse/skills/packs/egui
  • egui-app-architecture: Strict acyclic L0–L7 workspace crate layering (xtask/src/layers.rs), UI containment at L6+, panic-free/unsafe-free Clippy gates, and Cargo profile tuning for responsive dev builds and size-optimized Wasm bundles.
  • egui-app-runner: Running and debugging the native eframe desktop app, headless CLI, offscreen egui visual snapshots, authenticated loopback TCP control channel (--control), headless/bridged MCP servers, and Trunk Wasm builds.
  • egui-command-engine: Three-tier command taxonomy (Engine CommandSpecs, Shell/Tool UI commands, and CLI/JSON-RPC/MCP/TCP transports), declarative menu wiring, schema-driven egui dialogs, and adversarial input validation.
  • egui-cow-document-model: Pure serializable L1 domain models, sparse Arc copy-on-write chunk/tile storage with fallible allocation, zero-copy snapshot undo/redo history with selection target restoration, and compile-time complete content-addressed persistence (blake3 + Zstandard).
  • egui-testing-infrastructure: 7-tier verification architecture covering cargo-mutants mutation testing, egui_kittest headless UI harnesses, GPU test serialization (gpu_lock), pinned real-file corpus oracles, panic_hunt command fuzzing, performance budgets (xtask perf), and self-auditing scorecards (xtask scorecard).

packs/gpui — Rust gpui-kit Desktop Application Pack

Verified patterns and component workflows for building native Rust desktop applications with gpui-kit and GPUI:

npx skills add domesticmouse/skills/packs/gpui
  • gpui-kit: Window bootstrapping (Root and notification toasts), headless-testable view constructors, controlled Input/Textarea/Editor states, live TextView::markdown task-list toggling via WeakEntity, h_resizable split panes, segmented TabBars, and typed keyboard action dispatch.

packs/hardware — Embedded & Hardware Development Pack

Skills for electronic circuit design, PCB/PCBA workflows, Eurorack front panels, embedded systems, RTOS development, and board bring-up:

npx skills add domesticmouse/skills/packs/hardware

Recommended Companion Skills: Pair this pack with aklofas/kicad-happy (npx skills add aklofas/kicad-happy) for comprehensive KiCad schematic/PCB/Gerber analysis, footprint/symbol inspection, and fabrication verification.

  • eurorack-panel: Design, inspect, and generate mated Doepfer A-100 3U Eurorack front panels in KiCad (.kicad_pcb) from a main circuit PCB (generate_eurorack_panel.py), including footprint-to-bushing mechanical offset compensation, M3 mounting slots, silkscreen callouts, grounded F.Cu/B.Cu copper shield planes, DRC checks, and Gerber/Excellon drill exports.
  • nuttx-architecture: Architectural guide and development workflows for the Apache NuttX RTOS (nuttx) and user-space application suite (apps) repositories, including board porting (Renesas RA4M1 targets), device drivers, kernel subsystems, and NuttShell (nsh) applications.
  • skidl: Circuit design in Python with SKiDL for generating KiCad netlists (.net), PCBs, XML BOMs, hierarchical subcircuits, Electrical Rules Checking (ERC), and live-verified JLCPCB/LCSC PCBA BOM CSVs (jlcpcb_search.py).

Repository Structure

Default skills live under skills/, and optional skill groups live under packs/<pack-name>/:

├── skills/
│   └── <skill-name>/
│       ├── SKILL.md        # Required: YAML frontmatter (name, description) + instructions
│       ├── references/     # Optional: additional documentation loaded on demand
│       ├── scripts/        # Optional: helper scripts and utilities
│       └── resources/      # Optional: templates, components, and static assets
└── packs/
    └── <pack-name>/
        └── <skill-name>/
            ├── SKILL.md
            ├── references/
            └── scripts/

License

See LICENSE for details.

Disclaimer

This is not an officially supported Google product.

About

Agent Skills

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages