AI & ML

Agent skill for user-tools - invoke with $agent-user-tools

Inter-agent message bus with channels, queues, routing, reliability, and multiple communication patterns

Agent skill for architecture - invoke with $agent-architecture

Run Codex CLI, Claude Code, OpenCode, or Pi Coding Agent via background process for programmatic control.

MCP tools for loading, saving, and validating claude-flow configuration with path traversal protection and schema validation

Agent skill for goal-planner - invoke with $agent-goal-planner

Machine learning development agent for end-to-end ML workflows: data preprocessing, model training, evaluation, hyperparameter tuning, and deployment

Stream-JSON chaining for multi-agent pipelines, data transformation, and sequential workflows

Agent skill for data-ml-model - invoke with $agent-data-ml-model

Mathematical AI interpretability with sheaf cohomology, spectral analysis, causal inference, and hallucination prevention

Agent skill for release-swarm - invoke with $agent-release-swarm

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

Track Apple devices/AirTags via FindMy.app on macOS.

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

Send and receive iMessages/SMS via the imsg CLI on macOS.

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

List, configure, authenticate, call, and inspect MCP servers/tools with mcporter over HTTP or stdio.

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.

Manage Apple Notes via memo CLI: create, search, edit.

X/Twitter via xurl CLI: post, search, DM, media, v2 API.

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

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

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

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

Run 150+ AI apps via inference.sh CLI (infsh) — image generation, video creation, LLMs, search, 3D, social automation. Uses the terminal tool. Triggers: inference.sh, infsh, ai apps, flux, veo, image generation, video generation, seedream, seedance, tavily

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.

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.

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

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

Jailbreak LLMs: Parseltongue, GODMODE, ULTRAPLINIAN.

Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). Use for fast k-NN search, large-scale vector retrieval, or when you need pure similarity search without metadata. Best for high-performance applications.

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models. Supports multi-turn image chat, visual question answering, and instruction following. Use for vision-language chatbots or image understanding tasks. Best for conversational image analysis.

Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation workflows, or needing tight Megatron-LM integration for RL scaling.

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

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

Pitfalls, examples, and edge cases for Hermes Kanban workers. The lifecycle itself is auto-injected into every worker's system prompt as KANBAN_GUIDANCE (from agent/prompt_builder.py); this skill is what you load when you want deeper detail on specific scenarios.