venetanji

Generate images and videos using ComfyUI workflows via direct REST API. Use when asked to create images, edit photos, generate videos, or run Flux/LTX2/Wan workflows. Triggers on: generate an image, create a video, Flux2, ComfyUI, text-to-image, image-to-image, image-to-video, scene generation, TTS, voice clone, workflow.

Build a dialogue-first narrative video. Use when the user wants a "scene", "drama", "monologue", "dramatic clip", or a character acting through lines (not a music video). Spec authors the character, an ElevenLabs audio bed (dialogue + ambience + pauses) and a shot list whose start/duration aligns to the dialogue cues, then this skill renders flux anchors → LTX ia2v shots → vconcat + overlay full audio. Sibling to `music-video` (which is song-first and bar-aligned).

Generate a complete music video from a YAML spec — song via suno-mcp, per-scene audio-reactive video via comfyui LTX-2.3 ia2v, final concatenation with the clean song audio overlaid. Use when the user asks for a music video, long-form video synced to music, multi-scene video with a soundtrack, or anything that chains multiple ia2v clips into one output. Triggers on: "music video", "video to a song", "make a song and a video", "clip to this song", "visualizer", "lyric video".

Shared toolkit for building per-shot reference images (anchors) and keyframe sequences via flux2 i2i/i2i2/multi-reference. Use when any pipeline — music-video, drama-video, a one-off render — needs a scene-specific anchor, a character pose sheet, or first/last keyframes for a transition. Exposes a `generate_anchor` CLI + a prompting guide the operator should skim before writing prompts. All production-quality anchors on this deployment go through this skill.

Generate AI music with Suno via MCP. Use when creating songs with custom lyrics, instrumental tracks, or specific genres/moods. Triggers on "generate a song", "create music", "make a track", "Suno", "AI music", or any music generation request.

Generate production-ready storyboards and shot lists for PolyU Storyworld LTX-2.3 video workflows. Use when: creating a multi-shot scene for characters, planning a video sequence, generating per-shot i2v prompts from a narrative. Integrates with storyworld MCP for character context and reference images.

Agent entrypoint for the PolyU Storyworld: load character profiles and reference images directly from the GitHub repo and HuggingFace dataset, store them under story/characters/, and provide a simple CLI/API for other skills to call. Replaces mcporter-first workflow with a Git-first workflow.

Spawn and manage character agents for the PolyU Storyworld collaborative storytelling project. Use when: a student provides a character code (e.g. 6166r) and wants their character instantiated as an agent; asked to create/spawn a character session; routing messages between character agents and the co-narrator; generating images or video for a character using their reference image; setting up a new character workspace (SOUL.md, IDENTITY.md, AGENTS.md). NOT for: general image generation (use comfyui skill), GitHub/PR workflows (use github skill).

ComfyUI image/video/audio generation and delivery. Use for ANY request to generate, edit, or animate images or videos, or to produce speech audio. Handles the full flow: generate → download → present to user. Triggers on: "generate image", "make a video", "animate", "text to image", "comfyui", "flux", "ltx", "image from photo", "voice", "tts", "narrate", user sends a photo for editing/animation.

Use when reading or writing a running ERPNext/Frappe instance from the command line or scripts — looking up / listing / counting documents, creating or updating docs, submitting/cancelling, running query reports, calling whitelisted (incl. ERPNext factory) methods, attaching Drive URLs, or running server-side python (frappe.db.sql aggregates, atomic multi-step transactions) without the docker-exec + file-copy ceremony. Generic over any token-authenticated ERPNext instance.

Use when reading Xero accounting data from the command line or scripts — listing/looking up contacts, invoices, payments, bank transactions, manual journals; running reports (Trial Balance, Balance Sheet, P&L, BankSummary, Aged Receivables); downloading invoice PDFs or attachments; or managing the Xero OAuth2 connection (authorize, refresh, and verify you are connected to the correct organisation, not a trial/wrong tenant). Generic over any Xero org.

Core operating knowledge for Frappe Framework and ERPNext — the metadata/DocType data model, how to drive a running instance via bench CLI and the REST/RPC API, and the full module map (Accounting, Selling, Buying, Stock, Manufacturing, HR/Payroll, Projects, Assets, Support, Quality, Customization). Use whenever a task involves reading, creating, or reasoning about ERPNext data, calling the Frappe API, running bench, or knowing which DocType holds what. Distilled from docs.frappe.io.

Reference for managing an OpenClaw gateway — the multi-agent runtime distributed as the `openclaw` npm package. Use when the user asks about agents (list, add, configure), skills (how they load, where to put them, symlinks), sandbox (Docker shared container, binds, setupCommand), Discord bot routing (allowFrom, allowBots), gateway service (restart, logs, config validation), or any OpenClaw config edit. Install-specific details (bot roster, host paths, custom URLs) live in CLAUDE.md — this skill is about OpenClaw itself.