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 anymouschina/TapCanvas --skill seedance-20git clone --depth 1 https://github.com/anymouschina/TapCanvasWrote 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/anymouschina/tapcanvas/seedance-20)<a href="https://agentmods.dev/skills/anymouschina/tapcanvas/seedance-20"><img src="https://agentmods.dev/badge/skills/anymouschina/tapcanvas/seedance-20/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/anymouschina/tapcanvas/seedance-20"><img src="https://agentmods.dev/badge/skills/anymouschina/tapcanvas/seedance-20.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00181 | $0.04185 |
| Opus 5.5 | $0.00072 | $0.01674 |
| Sonnet 5.5 | $0.00036 | $0.00837 |
| Haiku 4.5 | $0.00018 | $0.00419 |
Grade A, and why
seedance-20 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 — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
seedance-20
Seedance 2.x operating loop for agent-directed video work. Use this root skill to select a verified model/surface profile, check facts, protect references, and keep prompts proportional to the requested duration before loading specialized sub-skills.
Soul
This skill exists so that a person who arrives with a feeling leaves with a film. Three principles govern everything below:
- Hear the intent behind the words. Users describe outcomes ("make it feel like home"), not parameters. Every gate and sub-skill translates feeling into craft; none of them may hand the translation work back to the user.
- Keep the story alive. Hold a story state across the conversation: subject, mode, look, references, decided constraints, and what failed before. Every skill reads it before asking anything and updates it after acting. A user should never have to repeat a decision, and a new request inherits the world already built.
- Evolve with the user. Speak plainly to a beginner and in director language to a professional - and notice when the same user grows from one into the other across a project. The register adapts; the standards never do.
Fast Lane
Most requests are one short clip from someone who just wants to see their idea. Do not run the full gate loop on them. Take the fast path when the request is a single standalone clip, from a non-expert, with no IP/likeness/brand/real-person or safety flag and no platform-fact question (API, pricing, model ID, limits, region):
- Go straight to
[skill:seedance-interview-short]- or write the brief immediately if the idea is already clear - then[skill:seedance-prompt-short]. - Apply craft inline from memory: one visible beat, one motivated camera move, one motivated light source, sound intent, and the directing coherence rule (name one intention; make camera, light, and performance serve it). Load
[ref:directing-engine],[ref:capability-map],[ref:allocation-model], and the source or professional gates only when something actually invokes them. - Treat it as one clip: do not ask sequence or continuation questions yet. Raise "should this be a series, part two, or longer" only after the first draft, or when the user says continue, extend, next part, or longer.
- Keep the single-clip prompt compact (about 40-110 words) unless the active surface is a verified stricter API, and keep director language (blocking, directorial voice, shot contracts) inside the internal brief - speak to the user in plain words.
What ships with it
60 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.
- docs/frontend-redesign.md 3.8 KB
- docs/QUICKSTART.es.md 4.5 KB
- docs/QUICKSTART.ja.md 5.5 KB
- docs/QUICKSTART.ko.md 4.9 KB
- docs/QUICKSTART.md 4.0 KB
- docs/QUICKSTART.ru.md 6.7 KB
- docs/QUICKSTART.zh.md 4.1 KB
- docs/README.ja.md 2.7 KB
- docs/README.ko.md 2.7 KB
- docs/README.zh.md 2.3 KB
- docs/RELEASE_v6.2.0.md 2.3 KB
- docs/RELEASE_v6.3.0.md 3.0 KB
- docs/RELEASE_v6.4.0.md 3.6 KB
- docs/RELEASE_v6.5.0.md 3.1 KB
- docs/RELEASE_v6.6.0.md 3.2 KB
- docs/v6-release-readiness.md 1.2 KB
- examples/golden-prompts/compact-i2v.md 834 B
- examples/golden-prompts/continuation-observed-deviation.md 967 B
- examples/golden-prompts/dense-2d-storyboard.md 1011 B
- examples/golden-prompts/first-last-frame-transition.md 897 B
- examples/golden-prompts/phased-single-take.md 862 B
- examples/golden-prompts/r2v-role-isolation.md 986 B
- examples/golden-prompts/sequence-continuation.md 1.0 KB
- examples/golden-prompts/video-edit-one-layer.md 843 B
- examples/sequence-airport-arrival/clip-01-contract.json 1003 B
- examples/sequence-airport-arrival/clip-01-prompt.md 608 B
- examples/sequence-airport-arrival/clip-01-take-review.json 786 B
- examples/sequence-airport-arrival/clip-02-continuation-contract.json 1.0 KB
- examples/sequence-airport-arrival/clip-02-prompt.md 725 B
- examples/sequence-airport-arrival/project-state.json 16 KB
- examples/sequence-airport-arrival/sequence-plan.md 1.7 KB
- examples/sequence-observed-deviation/project-state-after.json 6.2 KB
- examples/sequence-observed-deviation/project-state-before.json 7.0 KB
- examples/sequence-observed-deviation/take-review.json 717 B
- examples/standalone-clip/project-state.json 3.1 KB
- examples/standalone-clip/prompt.md 338 B
- LICENSE 1.1 KB
- README.md 41 KB
- references/2d-anime-grammar.md 4.7 KB
- references/agent-compatibility.md 9.2 KB
- references/allocation-model.md 3.5 KB
- references/anti-slop-lexicon.md 3.7 KB
- references/api-status.md 21 KB
- references/api-workflow.md 5.8 KB
- references/aspect-ratio-delivery.md 2.6 KB
- references/audio-guide.md 8.4 KB
- references/audio-post-delivery.md 2.2 KB
- references/awesome-seedance-import.md 1.3 KB
- references/capability-map.md 6.3 KB
- references/cinematography-shot-language.md 3.3 KB
- references/color-pipeline-aces.md 3.0 KB
- references/community-source-methodology.md 1.9 KB
- references/continuation-handoff.md 4.0 KB
- references/continuity-qc.md 1.2 KB
- references/delivery-qc.md 2.3 KB
- references/dense-storyboard-mode.md 3.0 KB
- references/directing-engine-genre-library.md 32 KB
- references/directing-engine.md 19 KB
- references/eval-rubric.md 1.2 KB
- references/event-density.md 1.0 KB
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 Changed · +23 lines 277bb2e225a1
- 1mo ago First seen · 118 lines · 181 tokens per session scan A 55b6f14a0a2a
seedance-20 is a skill published in the GitHub repository anymouschina/TapCanvas (709 stars, last pushed 4d ago), licensed MIT. It adds 181 tokens to every session and 4,185 once invoked, about $0.0007 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 2026-09-21.
seedance-20 compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| seedance-20anymouschina/TapCanvas | Skill | 709 | 4d ago | MIT | A | 181 | Claude Code |
| seedance-prompt-endexhunter/seedance2-skill | Skill | 4.4k | 1mo ago | MIT | A | 76 | Claude Code, Codex |
| seedance-prompt-zhdexhunter/seedance2-skill | Skill | 4.4k | 1mo ago | MIT | A | 100 | Claude Code, Codex |
| dramakexixihhhh/ai-short-drama-skill | Skill | 88 | 2mo ago | MIT | A | 279 | Claude Code, Codex |
| last30Dayschaitanyagiri/munder-difflin | Skill | 9.3k | 6d ago | MIT | A | 83 | Claude Code |
| thisWeekchaitanyagiri/munder-difflin | Skill | 9.3k | 6d ago | MIT | A | 77 | Claude Code |
| short-drama-storyboardzenstory-ai/drama-skills | Skill | 3.8k | yesterday | MIT | A | 102 | Codex |