IAMCCS CineInfo V2

IAMCCS_CineInfoV2 is the "breakout box" of the IAMCCS Cine pipeline: you feed it one cine_linx package - the single fat wire that carries everything between Cine stages - and it fans that out into seventeen individual, readable outputs. Guide plan, FLF timeline, global and per-segment prompts, segment lengths, dimensions, preview images, even the original linx for pass-through. It's the node you put in a graph when you want to see what the pipeline is planning, or when a downstream tool needs one specific value (like max_frames or promptrelay_epsilon) that's currently buried inside a JSON blob.

If you're new to this pack, think of cine_linx as a bus that the Cine SuperNodes pass between them - a single connection carrying what used to be a fan-out of a dozen wires. Nodes like IAMCCS_CineRelayOrBypass read their configuration straight off that bus. CineInfoV2 is the inverse: it's the tool for unpacking the bus when you need one of its fields in an explicit form.

How it works

Under the hood it takes the cine_linx package and rebuilds a verified guide plan for the FLF Productor: it parses the FLF timeline (via the same timeline parser the CineAllInOneFLFEngine uses), resolves each keyframe to a reference index, frame number, and strength, then exposes everything. Because it's a "production breakout," it's built to be trustworthy rather than pretty - the guide_plan_json and flf_timeline outputs are the exact structures the downstream generation actually consumes.

Inputs and outputs

One required input: cine_linx (type IAMCCS_SUPERNODE_LINX). No optional inputs.

Outputs, in rough groups:

• Plan: guide_plan_json, flf_timeline, global_prompt, local_prompts, segment_lengths, promptrelay_epsilon, report
• Numbers: max_frames, duration_seconds_int, frame_rate_int, width, height
• Images: multi_output (the full reference batch), image_1, first_stage_preview
• Pass-through: cine_linx (re-emitted unchanged), plus guide_data (a GUIDE_DATA-typed object for nodes that want the structured guide rather than the JSON string)

For a beginner, the two to reach for are guide_plan_json (paste it into a text viewer to see exactly what keyframes the pipeline computed) and first_stage_preview (a quick look at what the first stage produced without hunting through your output folder).

Installing it

Part of the IAMCCS-nodes pack. ComfyUI Manager → search IAMCCS, or:

cd ComfyUI/custom_nodes
git clone https://github.com/IAMCCS/IAMCCS-nodes.git

Restart ComfyUI; it's under IAMCCS/Cine/00 Utilities. No pip requirements, no model downloads - it only reads and reformats data already flowing through your graph. README floor: ComfyUI ≥ 0.3.0, Python ≥ 3.12, Torch ≥ 2.8.

Where people get burned

• It only works when fed a real cine_linx. If you're wiring this into a graph without the Cine SuperNodes, there's nothing to unpack - the input type is a custom IAMCCS_SUPERNODE_LINX, not a generic STRING. This is a debugging/breakout node for an existing Cine graph, not a starting point.
• With no other inputs, everything it outputs is derived from the linx you fed it. If the values look wrong, the problem is upstream in the ShotPlanner/planner that built the package, not in this node - and that's exactly what the report output helps you confirm.
• It's not an output node by default, so the seventeen outputs don't auto-preview. If you want to eyeball first_stage_preview, attach a preview node to it.