Skills seedance-v2

📦

seedance-v2

Name: seedance-v2
Author: runcomfy-com

Content revision r1 High Risk ⚙️ External commands🌐 Network access📁 Filesystem access

Generate Seedance Videos on RunComfy

Short video production needs clear model settings, reference media limits, and safe RunComfy commands. This skill guides Claude, Codex, and Claude Code through Seedance 2.0 Pro prompts and CLI inputs.

Supports: Claude Codex Code(CC)

⚠️ 38 Poor

Stop for confirmation before installing.

Review the plan and obtain explicit user consent before changing files.

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "seedance-v2" from https://skillstore.io/skills/runcomfy-com-seedance-v2.md and its manifest at https://skillstore.io/api/skills/runcomfy-com-seedance-v2/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/runcomfy-com-seedance-v2.md Signed manifest GET /api/skills/runcomfy-com-seedance-v2/manifest Signed lockfile GET /api/skills/runcomfy-com-seedance-v2/lockfile

Test it

Using "seedance-v2". A cafe owner wants a short vertical video of a barista speaking about a seasonal drink.

Expected outcome:

Recommended Seedance 2.0 Pro with 9:16 framing, 8 second duration, one character image, warm dialogue tone, and generate_audio enabled.

Using "seedance-v2". A film team wants a quick previs shot using a location video and a voice reference.

Expected outcome:

Prepared a multi-modal prompt that separates stable references from evolving action, camera movement, lighting, and audio direction.

Using "seedance-v2". A user asks for a 30 second branded clip from one prompt.

Expected outcome:

Explained the 15 second endpoint limit and suggested splitting the concept into shorter segments.

Security Audit

High Risk

v4 • 7/5/2026 Open versioned report

Most static command detections are Markdown backticks, schema examples, or prompt examples. Confirmed risks are the RunComfy and npm commands, the documented outbound Model API call, and a repository mismatch in the install instructions. No prompt injection attempt or hidden data exfiltration was found in SKILL.md.

Files scanned

182

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (1)

High

SKILL.md:20 SKILL.md:24-25

Repository Mismatch in Installation Instructions

SKILL.md links and installs agentspace-so/runcomfy-skills while this marketplace audit covers runcomfy-com. Users could install content outside the audited source.

The GitHub link and npx install command both name agentspace-so/runcomfy-skills. A benign repository move is possible, but the published install path differs from the audited source.

Capability review items (10)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Medium

SKILL.md:10

Ruby/shell backtick execution

Calls `runcomfy run bytedance/seedance-v2/pro` through the local

The skill states it calls the local runcomfy CLI, which executes an external program and submits data to RunComfy. This is intended functionality but still needs user consent and command review.

Medium

SKILL.md:24-26

Ruby/shell backtick execution

```bash

The fenced command runs npx skills add with global installation from a community repository. That can fetch and install unaudited code, so it is a real supply-chain command risk.

Medium

SKILL.md:46-47

Ruby/shell backtick execution

1. **RunComfy CLI** — `npm i -g @runcomfy/cli`

The prerequisite directs a global npm install of @runcomfy/cli. Installing global packages can execute third-party package lifecycle code and change local tooling.

Medium

SKILL.md:47-48

Ruby/shell backtick execution

2. **RunComfy account** — `runcomfy login` opens a browser device-code flow.

runcomfy login invokes an external CLI authentication flow and creates local credentials. It is normal setup, but it is still an external command with credential impact.

Medium

SKILL.md:70-74

Ruby/shell backtick execution

```bash

The bash example runs runcomfy with user-supplied input and an output directory. This is intended, but it executes a local CLI and writes generated files.

Medium

SKILL.md:78-87

Ruby/shell backtick execution

```bash

The bash example runs runcomfy with a prompt, reference URL, and output directory. This is intended, but it executes a local CLI and writes generated files.

Medium

SKILL.md:91-100

Ruby/shell backtick execution

```bash

The bash example runs runcomfy with image, video, and audio reference inputs. This is intended, but it executes a local CLI and sends user-provided media references.

Medium

SKILL.md:173

Ruby/shell backtick execution

The skill invokes `runcomfy run bytedance/seedance-v2/pro` with a JSON body matching the schema. The

This line confirms the CLI posts requests to the RunComfy Model API and downloads generated output. That network and filesystem behavior is central to the skill and must be disclosed.

Medium

SKILL.md:177

Ruby/shell backtick execution

- **Token storage**: `runcomfy login` writes the API token to `~/.config/runcomfy/token.json` with m

The line describes runcomfy login storing an API token under the user config directory. That makes the command sensitive even though the storage mode is documented.

Low

SKILL.md:173

Hardcoded URL

The skill invokes `runcomfy run bytedance/seedance-v2/pro` with a JSON body matching the schema. The

This line documents outbound requests to model-api.runcomfy.net and downloads from RunComfy domains. It is an intended external data transfer and output download path.

Risk Factors

⚙️ External commands (37)

🌐 Network access (8)

SKILL.md:14 SKILL.md:20 SKILL.md:82 SKILL.md:95 SKILL.md:96 SKILL.md:97 SKILL.md:169 SKILL.md:173

📁 Filesystem access (6)

SKILL.md:82 SKILL.md:95 SKILL.md:96 SKILL.md:97 SKILL.md:177 SKILL.md:177

Audited by: codex View Audit History →

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/runcomfy-com-seedance-v2/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/runcomfy-com-seedance-v2/security.svg)](https://skillstore.io/skills/runcomfy-com-seedance-v2?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/runcomfy-com-seedance-v2?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/runcomfy-com-seedance-v2/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/runcomfy-com-seedance-v2.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

runcomfy-com. (2026). seedance-v2 security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/runcomfy-com-seedance-v2/audits/4

BibTeX citation

@techreport{runcomfy-com-runcomfy-com-seedance-v2-2026,
  author = {runcomfy-com},
  title = {seedance-v2 security audit report (audit version 4)},
  institution = {Skillstore},
  year = {2026},
  number = {4},
  url = {https://skillstore.io/skills/runcomfy-com-seedance-v2/audits/4},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "seedance-v2 security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
  - name: "runcomfy-com"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/runcomfy-com-seedance-v2/audits/4"
identifiers:
  - type: other
    value: "skillstore:runcomfy-com-seedance-v2:audit:4"
    description: "Skillstore immutable audit report identifier"

Compare variants

3 installable variants

Each author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.

Why this variant is first

Highest Skillstore Score

Author / source Skillstore Score Evidence Confidence Skillstore usage Updated

doany-ai Recommended

doany-ai-seedance-v2

Skillstore Score 77

Evidence Confidence High

Skillstore usage 18

Updated

2026-08-21

agentspace-so

agentspace-so-seedance-v2

Skillstore Score 50

Evidence Confidence High

Skillstore usage 15

Updated

2026-08-21

runcomfy-com Current

runcomfy-com-seedance-v2

Skillstore Score 38

Evidence Confidence High

Skillstore usage 15

Updated

2026-08-21

Skillstore Score

Why this score Evidence Confidence: High

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Create Dialogue Ads

Generate short vertical spokesperson clips with reference images, guided audio tone, and clear duration settings.

Prototype Cinematic Shots

Turn shot descriptions, scene references, and motion notes into concise RunComfy prompts for previs work.

Test Brand Video Variants

Use fixed seeds, aspect ratios, and reference media to compare short-form video variants.

Try These Prompts

Create a Text-Only Clip

Generate a 5 second Seedance 2.0 Pro clip about [scene]. Use [aspect ratio], [camera motion], and [audio mood].

Create a Lip-Synced Ad

Use my character image to make an 8 second vertical ad. The speaker says [message] in a [tone] voice.

Plan Multi-Modal References

Prepare a Seedance prompt using my image, scene video, and audio reference. Keep identity stable and describe only action, lighting, and camera.

Compare Models Before Running

Compare Seedance 2.0 Pro with HappyHorse, Wan, Kling, and LTX for this request: [goal]. Recommend the best route and explain why.

Best Practices

Put stable identity and brand assets in image references instead of long text descriptions.
Keep prompts focused on action, camera, lighting, mood, and audio tone.
Confirm RunComfy commands and output directories before executing the CLI.

Avoid

Do not mix conflicting visual styles across reference media.
Do not request clips longer than the endpoint limit.
Do not submit sensitive faces, voices, or brand assets without permission.

Frequently Asked Questions

What does this skill help me create?

It helps create short Seedance 2.0 Pro videos through RunComfy with text, image, video, and audio references.

Does it run locally?

It uses a local RunComfy CLI, but generation runs through the RunComfy Model API.

What media references are supported?

The skill documents up to 9 images, 3 videos, and 3 audio references for this endpoint.

Can it make long videos?

No. The documented endpoint supports 4 to 15 second clips.

Can it create lip-synced speech?

Yes. It explains how to prompt for in-pass audio and natural lip-sync.

What should I review before using it?

Review the CLI command, output directory, RunComfy account status, and any sensitive media before submission.

Developer Details

Author

runcomfy-com

License

MIT

Skillstore revision

Version notice

The author did not declare a version.

Repository

https://github.com/runcomfy-com/skills/tree/main/seedance-v2/

Ref

62e2a730c5cd74eab4c7164309d810de660fcea3

Maintenance freshness

7/18/2026

Usage

8 downloads · 0 views

File structure

📄 SKILL.md