Create and edit Obsidian Bases (.base files) with views, filters, formulas, and summaries. Use when working with .base files, creating database-like views of notes, or when the user mentions Bases, table views, card views, filters, or formulas in Obsidian.

Create and edit Obsidian Flavored Markdown with wikilinks, embeds, callouts, properties, and other Obsidian-specific syntax. Use when working with .md files in Obsidian, or when the user mentions wikilinks, callouts, frontmatter, tags, embeds, or Obsidian notes.

>

>-

|

Adapt designs to work across different screen sizes, devices, contexts, or platforms. Ensures consistent experience across varied environments.

Review a feature and enhance it with purposeful animations, micro-interactions, and motion effects that improve usability and delight.

Improve layout, spacing, and visual rhythm. Fixes monotonous grids, inconsistent spacing, and weak visual hierarchy to create intentional compositions.

Perform comprehensive audit of interface quality across accessibility, performance, theming, and responsive design. Generates detailed report of issues with severity ratings and recommendations.

Amplify safe or boring designs to make them more visually interesting and stimulating. Increases impact while maintaining usability.

>

Improve unclear UX copy, error messages, microcopy, labels, and instructions. Makes interfaces easier to understand and use.

Add strategic color to features that are too monochromatic or lack visual interest. Makes interfaces more engaging and expressive.

Evaluate design effectiveness from a UX perspective. Assesses visual hierarchy, information architecture, emotional resonance, and overall design quality with actionable feedback.

Add moments of joy, personality, and unexpected touches that make interfaces memorable and enjoyable to use. Elevates functional to delightful.

Strip designs to their essence by removing unnecessary complexity. Great design is simple, powerful, and clean.

Extract and consolidate reusable components, design tokens, and patterns into your design system. Identifies opportunities for systematic reuse and enriches your component library.

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications. Generates creative, polished code that avoids generic AI aesthetics.

Create stunning, animation-rich HTML presentations from scratch or by converting PowerPoint files. Use when the user wants to build a presentation, convert a PPT/PPTX to web, or create slides for a talk/pitch. Helps non-designers discover their aesthetic through visual exploration rather than abstract choices.

Improve interface resilience through better error handling, i18n support, text overflow handling, and edge case management. Makes interfaces robust and production-ready.

Normalize design to match your design system and ensure consistency

Design or improve onboarding flows, empty states, and first-time user experiences. Helps users get started successfully and understand value quickly.

Improve interface performance across loading speed, rendering, animations, images, and bundle size. Makes experiences faster and smoother.

Push interfaces past conventional limits with technically ambitious implementations. Whether that's a shader, a 60fps virtual table, spring physics on a dialog, or scroll-driven reveals — make users ask "how did they do that?

Final quality pass before shipping. Fixes alignment, spacing, consistency, and detail issues that separate good from great.

Tone down overly bold or visually aggressive designs. Reduces intensity while maintaining design quality and impact.

Remove AI writing patterns from prose. Use when drafting, editing, or reviewing text to eliminate predictable AI tells.

One-time setup that gathers design context for your project and saves it to your AI config file. Run once to establish persistent design guidelines.

Improve typography by fixing font choices, hierarchy, sizing, weight consistency, and readability. Makes text feel intentional and polished.

Manage agent memory and user profile, and decide what belongs in the Memory OS vault. On Rogo there is no `memory` tool — use the Memory OS CLI (`~/.claude/memory-os/mos`). Activate when recording/recalling durable facts, deciding where a fact belongs, or managing memory budget.

Airtable REST API via curl. Records CRUD, filters, upserts.

Apple/macOS platform integrations — Notes, Reminders, Find My, iMessage, and macOS desktop automation. These skills are macOS-only. Use when the user mentions Apple Notes, Reminders, Find My / AirTag, iMessage / SMS, or wants to drive a Mac app (requires a desktop/computer_use tool; not available in Rogo unless configured).

Dark-themed SVG architecture/cloud/infra diagrams as HTML.

Search arXiv papers by keyword, author, category, or ID.

ASCII art: pyfiglet, cowsay, boxes, image-to-ascii.

ASCII video: convert video/audio to colored ASCII MP4/GIF.

AudioCraft: MusicGen text-to-music, AudioGen text-to-sound.

Axolotl: YAML LLM fine-tuning (LoRA, DPO, GRPO).

Article illustrations: type × style × palette consistency.

Knowledge comics (知识漫画): educational, biography, tutorial.

Infographics: 21 layouts x 21 styles (信息图, 可视化).

BlockTank BV accounting operations — bank reconciliation, P&L reconstruction, file organization, receipt extraction, and evidence location. Covers the full workflow from raw financial evidence to structured accounting data.

BlockTank Studio — "Ship Before They Ask" automation services (websites, ads, branding, video), Web3 video engine, cold-outreach pipeline. Project context, tool stack, and active initiatives.

Monitor blogs and RSS/Atom feeds via blogwatcher-cli tool.

>

Access bunq bank API via Python SDK (v1.28+) for transaction exports, account balances, and payment queries. Trigger: fetch bunq transactions, bunq API, bank exports

How to write chat messages (Telegram, Discord, Slack, etc.) without accidentally triggering parser-driven side effects — file deliveries, bot commands, mentions, or any other token the chat platform or agent runtime interprets as a live action. Triggers on any message composition that might include paths, commands, code blocks, or syntax with platform-specific meaning.

Use for any chat message where the user wants Claude Code CLI (Opus 4.8) to do the work. M3 (or any chat model) acts as a minimal-token passthrough: receive message, invoke the relay, return Claude's response. Designed for cost optimization when you have a flat-rate Claude subscription + a pay-per-token chat model. 30-90s latency per message.

Design one-off HTML artifacts (landing, deck, prototype).

OpenAI's model connecting vision and language. Enables zero-shot image classification, image-text matching, and cross-modal retrieval. Trained on 400M image-text pairs. Use for image search, content moderation, or vision-language tasks without fine-tuning. Best for general-purpose image understanding.

Inspect codebases w/ pygount: LOC, languages, ratios.

Coding agent CLI delegation — Anthropic's Claude Code, OpenAI's Codex, and OpenCode. Use when delegating a coding task to an external agent CLI (one-shot, worktree-isolated, or background interactive). Each agent has its own auth model, sandbox flags, and PTY requirements; pick the section that matches the agent.

Generate images, video, and audio with ComfyUI — install, launch, manage nodes/models, run workflows with parameter injection. Uses the official comfy-cli for lifecycle and direct REST/WebSocket API for execution.

Debug Hermes TUI slash commands: Python, gateway, Ink UI.

Author/validate/export Google's DESIGN.md token spec files.

Audit and reconcile documents across multiple sources and repositories. Covers cross-source file reconciliation (find duplicates, gaps, inconsistencies between directories) and repository completeness verification (ensure all expected documents are present and processed).

Exploratory QA of web apps: find bugs, evidence, reports.

DSPy: declarative LM programs, auto-optimize prompts, RAG.

Reconstruct and audit Dutch tax liabilities (VPB, LH, BTW) for Blocktank B.V. — multi-account patterns, betalingskenmerken, and MacBook recovery workflow.

Deep-dive bulk extraction of email attachments across multiple accounts, folders, and search strategies. Use when you need to harvest ALL documents from email for a project (e.g. tax doc collection, legal discovery, compliance).

lm-eval-harness: benchmark LLMs (MMLU, GSM8K, etc.).

Hand-drawn Excalidraw JSON diagrams (arch, flow, seq).

Find nearby places (restaurants, cafes, bars, pharmacies, etc.) using OpenStreetMap. Works with coordinates, addresses, cities, zip codes, or Telegram location pins. No API keys needed.

TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.

GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.

Search/download GIFs from Tenor via curl + jq.

GitHub credential management and file operations — PAT extraction, token recovery, and Contents API uploads. Use when GitHub auth fails, tokens need extraction from binary files, or when cloning is impractical and you need to upload files directly.

GitHub operations playbook — authentication setup, repository management, pull request lifecycle, code review, and issue triage. Covers `gh` CLI recipes with `git` + `curl` fallbacks for machines without `gh`. Use when the user mentions GitHub repos, PRs, issues, code review, or repo administration.

Jailbreak LLMs: Parseltongue, GODMODE, ULTRAPLINIAN.

Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration for Rogo. Uses Rogo-managed OAuth2 setup, prefers the Google Workspace CLI (`gws`) when available for broader API coverage, and falls back to the Python client libraries otherwise.

Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training

Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework

Configure, extend, or contribute to Hermes Agent.

Author in-repo SKILL.md: frontmatter, validator, structure.

HeartMuLa: Suno-like song generation from lyrics + tags.

Hermes infrastructure operations — inference provider configuration, API usage tracking, OAuth token migration, .env management, multi-agent setup, and system health audits. Covers the devops layer beneath Hermes agent itself.

Debug Python: pdb REPL + debugpy remote (DAP).

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

Pre-commit review: security scan, quality gates, auto-fix.

Write ML papers for NeurIPS/ICML/ICLR: design→submit.

Modify, debug, or extend the s6-overlay supervision tree inside the Hermes Agent Docker image — adding new services, debugging profile gateways, understanding the Architecture B main-program pattern.

SAM: zero-shot image segmentation via points, boxes, masks.

Rogo uses the Memory OS CLI (~/.claude/memory-os/mos) for session recall. This skill documents raw session file reading for reference. If needed for Rogo: read JSON transcript files in ~/.claude/projects/ to recover conversation history.

Parallel 3-agent cleanup of recent code changes.

Throwaway HTML mockups: 2-3 design variants to compare.

Audio spectrograms/features (mel, chroma, MFCC) via CLI.

Songwriting craft and Suno AI music prompts.

Throwaway experiments to validate an idea before build.

Spotify: play, search, queue, manage playlists and devices.

CLI to manage emails via IMAP/SMTP. Use himalaya to list, read, write, reply, forward, search, and organize emails from the terminal. Supports multiple accounts and message composition with MML (MIME Meta Language).

HuggingFace hf CLI: search/download/upload models, datasets.

Humanize text: strip AI-isms and add real voice.

Generate project ideas via creative constraints.

Iterative Python via live Jupyter kernel (hamelnb).

Hermes Kanban multi-agent workflow — decomposition playbook for orchestrators, pitfalls and examples for workers. **Note: Rogo does not have built-in kanban_* tools; this skill is a Hermes-specific guide.** Use only if Rogo has custom kanban integration configured.

Use when a Rogo Kanban worker wants to run Codex CLI as an isolated implementation lane while Rogo keeps ownership of task lifecycle, reconciliation, testing, and handoff (Hermes-specific — limited applicability to Rogo).

Linear: manage issues, projects, teams via GraphQL + curl.

llama.cpp local GGUF inference + HF Hub model discovery.

Karpathy's LLM Wiki: build/query interlinked markdown KB.

Manim CE animations: 3Blue1Brown math/algo videos.

Geocode, POIs, routes, timezones via OpenStreetMap/OSRM.

Use the mcporter CLI to list, configure, auth, and call MCP servers/tools directly (HTTP or stdio), including ad-hoc servers, config edits, and CLI/type generation.

Navigate and maintain the Memory OS v5 — Blocktank's structured memory platform across 7 content spaces (accounting, health, blocktank-studio, dms, tankie, jsburgos, molty) with semantic search and Rogo-owned scheduled execution. Activate on any domain or explicit request.

Host modded Minecraft servers (CurseForge, Modrinth).

Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.

Edit PDF text/typos/titles via nano-pdf CLI (NL prompts).

MCP client: connect servers, register tools (stdio/HTTP). Configure in Rogo's config for native MCP integration.

Find specific products available in the Netherlands, compare prices across NL retailers, and recommend the best option with delivery details.

Debug Node.js via --inspect + Chrome DevTools Protocol CLI.

Notion API + ntn CLI: pages, databases, markdown, Workers.

OBLITERATUS: abliterate LLM refusals (diff-in-means). Invoke via CLI (AGPL-3.0).

Read, search, create, and edit notes in the Obsidian vault.

Extract text from PDFs/scans (pymupdf, marker-pdf).

OCR and document extraction pipeline for financial/accounting documents. Covers PDF batch OCR (marker-pdf), screenshot batch OCR, image OCR, and document review. (Hermes path — N/A in Rogo)

OpenClaw ecosystem integration — multi-platform search (Agent Reach), task coordination (Paperclip), conversation recall, ZHS ZBot worker, and migration from OpenClaw to Hermes. Covers all OpenClaw-originated tools and workflows.

Control Philips Hue lights, scenes, rooms via OpenHue CLI.

Outlines: structured JSON/regex/Pydantic LLM generation.

p5.js sketches: gen art, shaders, interactive, 3D.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Pixel art w/ era palettes (NES, Game Boy, PICO-8).

Plan mode: write an actionable markdown plan to the active workspace's plans/ folder, no execution. Bite-sized tasks, exact paths, complete code.

Play Pokemon via headless emulator + RAM reads.

Query Polymarket: markets, prices, orderbooks, history.

54 real design systems (Stripe, Linear, Vercel) as HTML/CSS.

Create, read, edit .pptx decks, slides, notes, templates.

Use when building creative browser demos with @chenglou/pretext — DOM-free text layout for ASCII art, typographic flow around obstacles, text-as-geometry games, kinetic typography, and text-powered generative art. Produces single-file HTML demos by default.

vLLM: high-throughput LLM serving, OpenAI API, quantization.

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.

Execute plans via Agent-tool subagents (Rogo equivalent of delegate_task; 2-stage review).

4-phase root cause debugging: understand bugs before fixing.

Tankie's own infrastructure — hybrid ROG + MacBook architecture, SOUL.md, model routing, provider failover, self-configuration, and skill authoring. Activates on any infrastructure, architecture, or self-config topic.

Tax memo creation, review, and version management — audit financial memos against primary evidence, generate professional PDFs from markdown, and manage multi-version tax memo workflows for accountant delivery.

Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.

TDD: enforce RED-GREEN-REFACTOR, tests before code.

Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.

Unsloth: 2-5x faster LoRA/QLoRA fine-tuning, less VRAM.

Webhook subscriptions: event-driven agent runs.

W&B: log ML experiments, sweeps, model registry, dashboards.

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.

Write implementation plans: bite-sized tasks, paths, code.

WSL2 setup guides — SSH access from other machines and Obsidian native install via WSLg. Covers the two most common WSL2 infrastructure needs.

Interact with X/Twitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks (OAuth 2.0), mentions, and user lookups.

Interact with X/Twitter via xurl, the official X API CLI. Use for posting, replying, quoting, searching, timelines, mentions, likes, reposts, bookmarks, follows, DMs, media upload, and raw v2 endpoint access.

YouTube transcripts to summaries, threads, blogs.

Yuanbao (元宝) groups: @mention users, query info/members. NOTE: Requires Rogo's Yuanbao integration tools — not available unless Rogo is configured with yb_* tool implementations.

>

Watch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video.

Load ALL context for Roger Organic Video Mode (Roger the Editor) — the Common S3nse campaign, conveyor belts, rulings, tooling, and current state. Invoke at the start of any session touching Roger video work, the Common S3nse promo, the engine bake-off, or the Review Gate UI. Triggers: "roger video", "organic video", "common s3nse", "the belts", "bake-off", "ETHDam footage".

Load ALL context for Roger Tax Mode (quarterly BTW evidence collection → reconciliation → accountant packet, Blocktank BV). Invoke for anything touching quarterly taxes, BTW filing, Erik Neve's aangifte, the Q2/Q3 packet, collectors (bunq/Gmail/Nmbrs/Recibos), the calibration diff, or resuming the parked Tax Mode. Triggers: "tax mode", "BTW", "Q2 packet", "Q3 taxes", "Erik filing", "calibration diff", "collectors".

|

|

|

|

|

|

|

|

|

|

|

|

|

|

|

|

|

|

|

|

|