krafton-ai

Make channel-facing speak output sound natural and easy to read. Trigger on: preparing a Slack, Telegram, or Discord reply with speak; rewriting markdown-heavy or AI-sounding prose before delivery; or when a response feels too formal, too structured, or too obviously model-written for a chat conversation.

Practical browser automation guidance for KiraClaw browser MCP tasks. Trigger on: opening a downloaded HTML file, testing a localhost app, validating a web flow in the browser, interacting with a browser game, or when Playwright/browser actions keep timing out or failing.

Create or refine KiraClaw/Codex skills as concise SKILL.md packages. Trigger on: "make a skill", "create a skill", "update this skill", "seed skill", "default skill", or when the user wants reusable agent workflow instructions instead of one-off prompt text.

Create polished single-file HTML presentations matching the ARIA design system. Trigger on: "create a presentation", "make slides", "build a slide deck", "HTML presentation", or when the user wants to present information visually as a slide deck.

Use when the user asks how to connect or configure Slack, Telegram, or Discord in KiraClaw, including which token or settings field to fill in, what platform-side setup is required, and when an engine restart is needed.

Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks

Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.

Presentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks

Integrate a new text-to-speech model into vLLM-Omni from HuggingFace reference implementation through production-ready serving with streaming and CUDA graph acceleration. Use when adding a new TTS model, wiring stage separation for speech synthesis, enabling online voice generation serving, debugging TTS integration behavior, or building audio output pipelines.

Add a new diffusion model (text-to-image, text-to-video, image-to-video, text-to-audio, image editing) to vLLM-Omni, including Cache-DiT acceleration and parallelism support (TP, SP/USP, CFG-Parallel, HSDP). Use when integrating a new diffusion model, porting a diffusers pipeline or a custom model repo to vllm-omni, creating a new DiT transformer adapter, adding diffusion model support, or enabling multi-GPU parallelism and cache acceleration for an existing model.