Nodes/MKRShift_Nodes/TiXL Frame Plan
ComfyUI Node

TiXL Frame Plan

Telling a Tooll graph what to expect

By criskb·Created 7 months ago·Updated 5 months ago· 0
TiXL Frame Plan
    • tixl_frame_plan_json
    • manifest_line
    • summary_json
    asset_path
    transportfile
    source_kindtexture
    layer_nameMKRShift Layer
    graph_nameMKRShiftBridge
    blend_modeAlpha
    metadata_json
    transport_plan_json
    notes

    TiXL is Tooll - the open-source, node-based live visuals tool, formerly Tooll 2, now Tooll 3, and the pack's alias "tooll bridge" gives the game away. If you're routing ComfyUI frames into a Tooll composition, you need two sides to agree: this node writes down what you're sending, and its partner MKRTiXLImport reads what Tooll sends back. MKRTiXLFramePlan is the outbound spec.

    Everything about it is a description, not a transmission. You declare which asset, which transport, and which layer of which graph it belongs to, and the node emits a tixl_frame_plan_json plus a manifest_line that a Tooll-side add-on (in the repo's addons/tixl/ scaffold) is expected to consume. Think of it as the packing slip for a frame handoff.

    What you set

    • asset_path - the file or folder to push.
    • transport - file, ndi, spout, or osc. Note the real-world split again: Spout is Windows, NDI is network, OSC is control data not video. Pick what your Tooll instance actually listens on.
    • source_kind - texture, image_sequence, video, or mask.
    • layer_name and graph_name - which Tooll layer and composition this frame belongs to. Get the names right or the frame lands in the wrong place.
    • blend_mode - Alpha, Add, Screen, or Multiply, matching how Tooll should composite it.

    Optional extras: metadata_json (arbitrary payload), transport_plan_json (feed a plan from the network plan nodes so transport details stay in sync), and notes.

    Outputs

    tixl_frame_plan_json (the full spec), manifest_line (compact handoff), and summary_json. Wire the plan into whatever consumes it on the Tooll side, or keep it alongside the TCP/Spout plans as part of one transport corner of your graph.

    Installing it

    In criskb/MKRShift_Nodes from Cris K B. ComfyUI Manager search "MKRShift_Nodes", or:

    cd ComfyUI/custom_nodes
    git clone https://github.com/criskb/MKRShift_Nodes
    

    Restart. No pip requirements, no models - the node just assembles JSON.

    The reality check

    Like the Spout and Syphon plan nodes, this is a contract, not a sender. There's no runtime Tooll upload node in the pack - the actual frame delivery is meant to go through the TiXL add-on scaffold that ships alongside. It's genuinely useful as the single source of truth for "which frame goes where," and it pairs cleanly with MKRTiXLImport for a round-trip. Just don't queue the graph expecting pixels to fly into Tooll; wire the plan to the add-on that does the flying.

    CategoryMKRShift Nodes/Addons/TiXL

    Inputs (9)

    NameTypeDefaultDescription
    asset_pathSTRING
    transportCOMBOfile4 options: file, ndi, spout, osc
    source_kindCOMBOtexture4 options: texture, image_sequence, video, mask
    layer_nameSTRINGMKRShift Layer
    graph_nameSTRINGMKRShiftBridge
    blend_modeCOMBOAlpha4 options: Alpha, Add, Screen, Multiply
    metadata_jsonoptSTRING
    transport_plan_jsonoptSTRING
    notesoptSTRING

    Outputs (3)

    NameTypeDescription
    tixl_frame_plan_jsonSTRING
    manifest_lineSTRING
    summary_jsonSTRING