agentsclimarketplace

Handoff

Skill wassermanproductions/master-canvas-hermes-plugin/skills/handoff

Hermes Agent plugin and bundled skill for Master Canvas AI video handoff packages.

Install
npx -y skills add wassermanproductions/master-canvas-hermes-plugin --skill handoff

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Work with Master Canvas AI video handoff packages. Use when a user provides a Master Canvas ZIP or project_manifest.json and wants Hermes to inspect assets, preserve scene and shot order, prepare ComfyUI/LTX 2.3 jobs, or hand off prompts to Kling, Veo, or another video generation operator. Also use when a user wants Hermes to autonomously create, update, and package a Master Canvas pre-production project from a brief, scenes, shots, prompts, references, and local assets.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

4.5 KB, as published. Nobody here has run it

Master Canvas Handoff

Use this skill when a user gives Hermes a Master Canvas handoff ZIP, extracted package folder, or project_manifest.json and wants video-generation work planned or executed from that package.

Also use it when the user wants Hermes to operate Master Canvas autonomously. In that mode, Hermes should create and maintain a Master Canvas-compatible package directly: scenes, shots, prompts, negative prompts, copied assets, references, ComfyUI/LTX jobs, Kling/Veo prompt sheets, and deliverable bins.

Default Workflow

  1. Inspect the package with mastercanvas_inspect_package.
  2. If the package is a ZIP and generation work will happen locally, extract it with mastercanvas_extract_package.
  3. Build the ComfyUI/LTX plan with mastercanvas_comfy_plan.
  4. Treat project_manifest.json as the source of truth for scene order, shot order, prompts, negative prompts, references, and output bins.
  5. Keep outputs organized by scene using deliverables/bin_plan.json or the outputBin fields in the plan.

Autonomous Master Canvas Workflow

When the user wants Hermes to create or manage a Master Canvas project:

  1. Use mastercanvas_create_package to create the package from the brief, scene list, shot list, prompts, and local asset paths.
  2. Use mastercanvas_upsert_scene when the user changes the scene structure, scene description, style prompt, or music prompt.
  3. Use mastercanvas_upsert_shot when the user adds or changes a shot, source image, prompt, negative prompt, camera direction, lens, lighting, duration, or notes.
  4. Use mastercanvas_inspect_package after changes to confirm scene count, shot count, missing assets, and readiness.
  5. Use mastercanvas_package_zip when the package is ready to share, archive, or send into downstream generation.

In autonomous mode, keep the package compatible with the Master Canvas app handoff contract. Do not invent a separate schema unless the user explicitly asks for a different format.

Package Contract

Expected Master Canvas handoff packages usually contain:

  • project_manifest.json: complete project context, scene list, shot list, asset list, prompt fields, negative prompt fields, and generation targets.
  • assets/: source images, uploaded video references, uploaded music/audio references, and other media.
  • comfyui/jobs/: one JSON job per shot when exported by Master Canvas.
  • kling-veo/: prompt sheets for operators using hosted generation tools.
  • hermes-agent/: agent-facing brief and context notes.
  • deliverables/bin_plan.json: target folder/bin structure for completed renders.

If one of these folders is missing, continue with the manifest and explain the limitation.

ComfyUI / LTX 2.3 Execution Guidance

When the user wants ComfyUI/LTX:

  • Preserve the shot sequence exactly.
  • Use the image path from sourceImage as the first-frame/image reference for each shot.
  • Use the provided prompt and negative prompt without dropping lens, lighting, action, continuity, or sound notes.
  • Target at least 1080p unless the manifest specifies a higher floor.
  • Generate each shot into its scene bin.
  • Return an inventory of completed shots grouped by scene.
  • Flag missing assets, impossible prompt requirements, or continuity conflicts before starting expensive generation.

Kling / Veo Operator Guidance

When the user wants Kling, Veo, or a human operator handoff:

  • Use the Kling/Veo prompt sheet if present.
  • Otherwise build a simple table with scene, shot order, source asset path, prompt, negative prompt, duration, and notes.
  • Keep file names and output bins stable so rendered clips can be reconciled against the original canvas.

Verification

Before reporting completion:

  • Confirm total scenes and shots.
  • Confirm whether any shots are missing source images.
  • Confirm where extracted assets or output bins are located.
  • Summarize the next concrete generation step.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.