← 返回 Skills 市场

IMA Seedance2.0 Video Generator

Name: IMA Seedance2.0 Video Generator
Author: allenfancy-gan

作者 allenfancy-gan · GitHub ↗ · v1.0.3 · MIT-0

cross-platform ✓ 安全检测通过

189

总下载

当前安装

版本数

在 OpenClaw 中安装

/install ima-seedance2-video-generator

功能描述

Seedance 2.0 AI video generator — two models in one skill: Seedance 2.0 (ima-pro) for cinema-grade quality with high frame-rate temporal consistency, precise...

使用说明 (SKILL.md)

IMA Seedance 2.0 — OpenClaw Calling Protocol

Hard Rule

OpenClaw must call scripts/ima_video_create.py.

OpenClaw must not:

build /open/v1/tasks/create payloads
compute attribute_id
compute credit
build src_img_url
build src_image
build src_video
build src_audio

Required Read Order

Supported User Intents

text-to-video
image-to-video
first-last-frame transition
reference-image-to-video
multimodal reference-image-to-video with image / video / audio inputs

Input Entry Points

--prompt
--model-id
--task-type when explicit task type is required
--input-images
--reference-image
--reference-video
--reference-audio
--extra-params

Stop Conditions

Stop before task creation if:

prompt is missing
model cannot be resolved
reference media preflight validation fails
any reference media compliance verification fails
create-task returns a hard failure

Task Type Rules

Input pattern	task_type
text only	`text_to_video`
one image	`image_to_video`
explicit first-last-frame with 2 images	`first_last_frame_to_video`
any video input	`reference_image_to_video`
any audio input	`reference_image_to_video`
multiple images without explicit first-last-frame intent	`reference_image_to_video`

Model ID Reference (CRITICAL)

Use exact model_id from this table. Do NOT infer from friendly names.

Friendly Name	model_id	Notes
Seedance 2.0	`ima-pro`	✅ Quality priority, 300~900s
Seedance 2.0 Fast	`ima-pro-fast`	⚠️ Speed priority, 120~600s

User input aliases:

Quality/Professional/Pro/专业版/高质量 → ima-pro
Fast/Speed/Quick/极速/快速 → ima-pro-fast
Default/默认 → ima-pro

Model Selection Priority

User preference (if explicitly stated) → highest priority
Fallback default: ima-pro

Task	Default	Fast Alternative
`text_to_video`	`ima-pro`	`ima-pro-fast`
`image_to_video`	`ima-pro`	`ima-pro-fast`
`first_last_frame_to_video`	`ima-pro`	`ima-pro-fast`
`reference_image_to_video`	`ima-pro`	`ima-pro-fast`

Minimal Invocation Examples

# Text to video
python3 {baseDir}/scripts/ima_video_create.py \
  --prompt "a puppy runs across a sunny meadow, cinematic"

# Single image
python3 {baseDir}/scripts/ima_video_create.py \
  --prompt "camera slowly zooms in" \
  --input-images https://example.com/photo.jpg \
  --model-id ima-pro-fast

# Explicit first-last-frame
python3 {baseDir}/scripts/ima_video_create.py \
  --task-type first_last_frame_to_video \
  --prompt "smooth transition" \
  --input-images https://example.com/first.jpg https://example.com/last.jpg

# Multimodal reference mode
python3 {baseDir}/scripts/ima_video_create.py \
  --reference-image https://example.com/product.jpg \
  --reference-video https://example.com/clip.mp4 \
  --reference-audio https://example.com/narration.mp3 \
  --model-id ima-pro-fast

References

Execution protocol: references/protocols/execution.md
Event stream: references/protocols/event-stream.md
Create-task contract: references/contracts/create-task.md
Credit rules: references/contracts/credit-rules.md
Payload rules: references/contracts/payload-rules.md
Reference-image-to-video flow: references/flows/reference-image-to-video.md
Text-to-video flow: references/flows/text-to-video.md
Reference media limits: references/limits/reference-media-rules.md
FAQ: references/support/faq.md
Troubleshooting: references/support/troubleshooting.md

安全使用建议

This package appears coherent for its stated purpose. Before installing, consider: (1) Use a scoped/test IMA_API_KEY (not a broad-production key) during initial testing because the key will be sent to the declared upload host (imapi.liveme.com) for local/non-HTTPS uploads. (2) If you run the skill non-interactively, IMA_AUTO_CONSENT can bypass compliance prompts — avoid setting it unless intentional. (3) The skill writes logs and prefs under ~/.openclaw; inspect or sandbox if you want to prevent local files being created. (4) Review SECURITY.md and the upload/verification code if you handle sensitive media (PII) because local media may be uploaded to the provider's storage/CDN. (5) If you want stricter control, run the scripts in a constrained environment (container/VM) and revoke or scope the API key after testing.

功能分析

Type: OpenClaw Skill Name: ima-seedance2-video-generator Version: 1.0.3 The IMA Seedance 2.0 Video Generator is a well-structured and documented skill bundle for generating AI videos. The codebase (primarily scripts/ima_video_create.py and its supporting modules) implements a complete workflow including media upload, compliance verification, and task polling. While the skill requires an IMA_API_KEY and communicates with external endpoints (api.imastudio.com and imapi.liveme.com), these behaviors are clearly disclosed in the SKILL.md and SECURITY.md files as necessary for the service's operation. The use of subprocess for ffmpeg/ffprobe is handled safely for media metadata extraction, and the local persistence is limited to user preferences and auto-cleaned logs.

能力评估

✓ Purpose & Capability

Name/description (Seedance 2.0 video generation) align with requested binaries (python3, ffmpeg, ffprobe), the single required env var (IMA_API_KEY), and the included Python scripts for task creation, uploads, and compliance. The included APP_ID/APP_KEY are used for upload signing and are documented as non-secret in SECURITY.md.

ℹ Instruction Scope

Runtime instructions require calling scripts/ima_video_create.py and the shipped scripts perform media validation, optional remote-URL probing, local-file uploads to the declared upload service (imapi.liveme.com), compliance verification prompts, and polling of api.imastudio.com. These actions are within the declared scope but the skill will (a) send your IMA_API_KEY to the upload host for local/non-HTTPS uploads and (b) may fetch user-supplied remote media URLs for validation. The skill respects interactive consent but supports non-interactive auto-consent via IMA_AUTO_CONSENT.

✓ Install Mechanism

No install spec (instruction-only) and only a simple requirements.txt (requests>=2.25.0). No arbitrary downloads or archive extraction are present in the package metadata. This is low-risk from an install mechanism perspective.

✓ Credentials

Only one required credential (IMA_API_KEY) is requested and it is justified by the skill's remote APIs. Optional env vars (IMA_DEBUG, IMA_AUTO_CONSENT, IMA_STDOUT_MODE) are documented. The code does not request unrelated cloud credentials or system secrets. Note: the IMA API key is sent to both api.imastudio.com and imapi.liveme.com (documented) — use a scoped/test key if you are concerned.

ℹ Persistence & Privilege

Persistence is limited to ~ /.openclaw/memory/ima_prefs.json and ~/.openclaw/logs/ima_skills/ (declared). always:false (not force-installed). The skill can run autonomously by default (platform default) and will use the provided IMA_API_KEY during autonomous calls — if you allow autonomous invocation, that key can be used without interactive prompts unless the skill encounters compliance checks (which may require consent or IMA_AUTO_CONSENT).

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install ima-seedance2-video-generator
安装完成后，直接呼叫该 Skill 的名称或使用 /ima-seedance2-video-generator 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.3

- Updated environment variable and credential documentation for clarity and conciseness. - Minor adjustments to metadata structure for consistency with current standards. - Added explicit `requests` dependency version specification. - Cleaned up envVars (removed optional BASE_URL overrides and condensed descriptions). - No changes to functionality or user-facing behavior.

v1.0.2

- Added ffmpeg and ffprobe as required runtime dependencies. - Expanded documentation for required and optional environment variables, including support for custom API endpoints. - Credential note now clarifies endpoint override behavior when IMA_BASE_URL or IMA_IM_BASE_URL are set. - No changes to core logic or functionality; documentation and environment metadata improved for clarity and integration.

v1.0.1

- Removed documentation and utility files: CHANGELOG.md, README.md, references/agent-guide.md, scripts/build_pack.sh, scripts/refactor_modularize.py, and scripts/setup.py. - No user-facing functionality changes; all core video generation capabilities remain unchanged. - Skill documentation and calling protocol are preserved in SKILL.md.

v1.0.0

IMA Seedance 2.0 Video Generator – Initial release - Introduces Seedance 2.0 AI video generation with both quality-focused ("ima-pro") and speed-focused ("ima-pro-fast") models. - Supports text-to-video, image-to-video, first-last-frame transitions, and multimodal (image/video/audio) reference-driven video generation. - Precise camera language control, cinematic multi-shot workflows, and consistent character animation included. - 2K video output and batch rapid prototyping capabilities. - Requires IMA_API_KEY for operation.

元数据

Slug ima-seedance2-video-generator

版本 1.0.3

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 4

常见问题