diff --git a/h3-plan-gate/SKILL.md b/h3-plan-gate/SKILL.md index 9093b9c..553b75d 100755 --- a/h3-plan-gate/SKILL.md +++ b/h3-plan-gate/SKILL.md @@ -224,6 +224,15 @@ system prompt emits. Report the delivered frames/seconds it prints. ## Step 5 — Deliver (the node makes the multi-clip video) +**This plan targets the standalone MiniMax H3 Context Loop pack only.** ComfyUI-Easy-Media +(yolain) ships its own H3 node family built on the MultiTrack system and has NO +`plan_json`/`shots` consumer — a chain plan pasted into an Easy-Media workflow is read by +no node, and the workflow's default single task segment (0-240 frames = exactly 10 s at +24 fps) renders the whole story as one 10-second clip. For that pack use the +`easy-media-track-gate` skill instead: one task segment per shot in the multiTrackEditor +TRACK_DATA, a for-loop rendering each segment as its own H3 generation, and +mergeVideosFromPaths stitching the clips. + Verified against the node pack source (`chain_nodes.py`, `MiniMaxH3ChainPlan` / `MiniMaxH3ChainPlanModern`): the parser is a plain `json.loads` over the effective plan string — a non-empty STRING connected to the forced `plan_json_input` socket