Camera Angle Selector (nikosis)

Naming a physical camera position is one of the highest-leverage things you can put in a prompt - it's the kind of concrete, photographic vocabulary that consistently outperforms vague quality words, on both older CLIP-based models and the newer LLM-encoded ones. This node is a one-dropdown shortcut for that trick: instead of remembering and typing "aerial view" or "medium shot" every time, you pick it from a list.

Fair warning up front: this node isn't actually covered in the pack's own README, which only documents the Aspect Ratio, Styles Selector, and Text Concatenate nodes. Everything below is grounded in what its info_schema actually reports plus the conventions the rest of the pack follows - treat the in-app node as the ground truth over any older doc.

What it does and why it's worth having

Camera and shot-framing language is one of the few prompting habits that got more valuable across the shift from CLIP-based models to LLM-encoded ones, not less - where quality-tag spam like "masterpiece, 8k, ultra detailed" has gone dead weight on newer models, naming an actual shot type or camera position is a concrete instruction the encoder can act on regardless of architecture. This node exists to make that easy: one dropdown, a curated list of camera-framing phrases, done. It's the single-purpose sibling to this pack's Styles Selector - same idea, narrower scope, and it clearly follows the same Category | Specific naming convention (Camera | Wide-Angle, Camera | Aerial View, and so on).

The one input, and the output

• angle - a single dropdown, defaulting to No Style (meaning: don't inject anything). The list is long - over 200 entries - and from what's visible it spans shot distance (Camera | Far-Shot, Camera | Medium-Shot) and elevation/perspective (Camera | Aerial View, Camera | Drone Footage, Camera | Satellite View). Scroll through it for the specific framing you want rather than guessing at the exact wording.
• positive_prompt (output) - the selected angle's phrase, as a plain string. There's no negative output here, unlike the Styles Selector - camera framing is additive by nature; you're not usually trying to steer away from a shot type, you're picking the one you want.

Wire positive_prompt into this pack's Text Concatenate node alongside your subject description and any style fragments, or straight into a CLIP Text Encode if camera framing is all you're adding on top of hand-typed text.

Installing it

Same pack, same install: ComfyUI Manager (search "ComfyUI-Nikosis-Nodes") or git clone https://github.com/Nikosis/ComfyUI-Nikosis-Nodes into ComfyUI/custom_nodes, pip install -r requirements.txt, restart. Nothing heavy - no models, minimal dependencies, same as the rest of this pack.

Things to watch for

Leaving it on "No Style" and wondering why nothing changed. The default is off. If you want the effect, you have to actually pick an entry from the dropdown.

Expecting a negative-steering option. There isn't one on this node - if you want to actively avoid a certain framing, that's a job for your own negative prompt on a CLIP Text Encode, not this node. And if you're running a guidance-distilled model at CFG 1 (Z-Image Turbo and similar), remember the negative prompt field does nothing there regardless of what you put in it - restate the framing you do want positively instead of trying to negate the one you don't.

No README to fall back on. Since this node isn't documented in the pack itself, if something about its behavior looks off, the fastest way to check is opening nodes/ in the installed pack folder and reading the source directly, rather than searching for docs that don't exist yet.