🔶 Main Prompt / Prompt Text

This is the base prompt node for the whole Chaosaiart animation system - the "default" positive and negative text that applies unless a frame-specific override kicks in later. If you're only using one prompt for your whole generation, this is functionally a fancier CLIPTextEncode pair bundled into one node with a couple of extra conveniences; if you're doing the pack's frame-by-frame prompt-changing trick, this is the fallback every chaosaiart_Prompt_Frame node measures itself against.

positiv and negativ are your standard multiline text fields - write your base prompt and negative prompt here exactly like you would anywhere else. The optional add_positiv and add_negativ fields let you append extra text without editing your main prompt directly, which is handy if you want to keep a clean "core" prompt and layer situational additions on top (a style tag, an extra constraint) without scrolling through a wall of text every time you tweak something. The optional add_lora input (type LORA) is where you chain in a LoRA built with chaosaiart_lora or chaosaiart_lora_advanced - this node doesn't apply the LoRA itself, it just carries the reference along inside its bundled output so something downstream can apply it.

That bundled output is MAIN_PROMPT, a custom type specific to this pack that packages your positive text, negative text, and LoRA chain together in one wire. It's meant to plug into chaosaiart_MainPromptCLIPEncode, which is the node that actually applies the LoRA and encodes the text into conditioning, or into chaosaiart_Prompt_mixer_byFrame, which uses it as the fallback prompt for frames that haven't hit a chaosaiart_Prompt_Frame override yet. The other three outputs - LORA, POSITIV_TXT, and NEGATIV_TXT - expose the same pieces individually as plain text/LoRA types, in case you want to route them to standard ComfyUI nodes instead of staying inside the 🔶Chaosaiart ecosystem for that part of your graph.

Install: search "Chaosaiart-Nodes" in ComfyUI Manager, or use "Install via Git URL" with https://github.com/chaosaiart/Chaosaiart-Nodes. By hand: cd ComfyUI/custom_nodes && git clone https://github.com/chaosaiart/Chaosaiart-Nodes, then restart ComfyUI. Linux users also need pip install opencv-python and pip install tqdm inside their venv - the pack's Windows installer runs this step for you automatically, the Linux path doesn't.

One thing worth knowing before you build a big graph around this: MAIN_PROMPT is a type this pack invents for itself, so it only talks to other 🔶Chaosaiart nodes downstream - you can't wire it straight into a stock CLIPTextEncode or another pack's prompt node. If you want to mix this into a non-Chaosaiart pipeline, use the plain POSITIV_TXT/NEGATIV_TXT string outputs instead and encode them the normal way.