Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add SlavaSexton/ComfyUI-Agent-Kit --skill seedancegit clone --depth 1 https://github.com/SlavaSexton/ComfyUI-Agent-KitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/slavasexton/comfyui-agent-kit/seedance)<a href="https://agentmods.dev/skills/slavasexton/comfyui-agent-kit/seedance"><img src="https://agentmods.dev/badge/skills/slavasexton/comfyui-agent-kit/seedance/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/slavasexton/comfyui-agent-kit/seedance"><img src="https://agentmods.dev/badge/skills/slavasexton/comfyui-agent-kit/seedance.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00105 | $0.02268 |
| Opus 5.5 | $0.00042 | $0.00907 |
| Sonnet 5.5 | $0.00021 | $0.00454 |
| Haiku 4.5 | $0.00011 | $0.00227 |
Grade A, and why
seedance scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Seedance video prompting
ByteDance's Seedance family generates video and audio jointly and takes multimodal references (images, video clips, audio clips) that a prompt addresses by label. Getting the labels and the task type right matters more than adjectives: the same sentence can be read as "reference this" or "edit this" depending on one word.
Load reference.md for the full formulas, the failure table, worked examples and the ComfyUI
parameters. Load supplementary.md only for non-official leads about 2.5; it never overrides
reference.md and carries a conflict ledger. This file is the router and the rules you need before
writing a single line.
Which model you are talking to
| Model | Where it runs | Refs per pass | Max length | In ComfyUI |
|---|---|---|---|---|
| Seedance 2.5 | Dreamina, Jimeng AI, Doubao Pro, and ComfyUI since core v0.31.0 | 30 img + 10 video + 10 audio (audio-only allowed) | 4 to 30s; nest-extend to 60s; Long Video mode 30 to 180s in one shot | Yes, since 2026-08-08 |
| Seedance 2.0 / 2.0 Mini / 2.0 Fast | BytePlus ModelArk, ComfyUI | 9 img + 3 video + 3 audio | 15s | Yes |
| Seedance 1.5 Pro | BytePlus ModelArk, ComfyUI | n/a (no omni-reference) | 4 to 12s, min 4s enforced | Yes, and the only one there with generate_audio |
| Seedance 1.0 Lite / Pro / Pro Fast | BytePlus ModelArk, ComfyUI | n/a | n/a | Yes |
Seedance 2.5 reached ComfyUI on 2026-08-08 (core v0.31.0, PR 15395), and this file said the opposite until 2026-08-09. The old line, "Seedance 2.5 has no ComfyUI nodes", was correct on 2026-08-01 and went stale a week later. Kept as a correction rather than silently replaced.
How to build it now. 2.5 is a model option inside the SAME ByteDance2* nodes, not new nodes:
ByteDance2TextToVideoNode, ByteDance2FirstLastFrameNode, ByteDance2ReferenceNode (category
partner/video/ByteDance). Set the model DynamicCombo to Seedance 2.5 and the widgets underneath
change. Minimal graph: the node's VIDEO output straight into SaveVideo.video.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 138 lines · 105 tokens per session scan A c33e6104d381
seedance is a skill published in the GitHub repository SlavaSexton/ComfyUI-Agent-Kit (109 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 105 tokens to every session and 2,268 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
cinematic-frame-director
Turn a user's idea, scene description, or mood into ONE production-grade cinematic still-image prompt in English for text-to-image models (Seedream, Midjourney, Flux, DALL-E, or any image generator). Use this skill whenever the user asks for an image prompt, a cinematic frame, a "movie still" look, a photorealistic…
prompt-refiner
Lints an existing user-supplied prompt for prompt-engineering defects and prompt-injection risks. Produces a diagnostic report with rule IDs, severity, line/column, quoted evidence, rationale, and optional JSON output. Does not rewrite prompts or impose prompt frameworks.
outlines
Outlines: structured JSON/regex/Pydantic LLM generation.
instructor
Structured LLM outputs validated with Pydantic.
guidance
Constrain LLM output with grammars; guarantee valid JSON.
dspy
DSPy: declarative LM programs, auto-optimize prompts, RAG.