← 返回 Skills 市场

Filtrix Video Gen

Name: Filtrix Video Gen
Author: lumenclaw-cloud

作者 lumenclaw-cloud · GitHub ↗ · v0.1.0 · MIT-0

cross-platform ⚠ suspicious

290

总下载

当前安装

版本数

在 OpenClaw 中安装

/install filtrix-video-gen

功能描述

Generate videos through Filtrix Remote MCP. Use when users ask for text-to-video, image-to-video, video task polling, or downloading completed videos with ge...

使用说明 (SKILL.md)

Filtrix Video Gen (MCP)

This skill is MCP-only.

Endpoint: https://mcp.filtrix.ai/mcp
Auth: Authorization: Bearer \x3CFILTRIX_MCP_API_KEY>
Primary tools:
- generate_video_text
- generate_video_image
- get_video_status

Setup

Required:

FILTRIX_MCP_API_KEY

Optional:

FILTRIX_MCP_URL (default: https://mcp.filtrix.ai/mcp)

Generate Video (Text)

python scripts/generate.py \
  --mode text-to-video \
  --prompt "a cinematic drone shot over a neon city at night" \
  [--aspect-ratio 16:9] \
  [--idempotency-key KEY] \
  [--wait] \
  [--poll-interval 8] \
  [--timeout 600] \
  [--output /tmp/video.mp4]

Default behavior submits a request and prints request_id. Add --wait to poll until completion and download the final video.

Generate Video (Image-to-Video)

Grok Imagine:

python scripts/generate.py \
  --mode grok-imagine \
  --prompt "camera slowly pushes in, fog drifting" \
  --image-path /path/to/input.png \
  --duration-seconds 6 \
  [--aspect-ratio 16:9] \
  [--wait]

Seedance 1.5 Pro:

python scripts/generate.py \
  --mode seedance-1-5-pro \
  --prompt "soft cinematic motion, subject turns to camera" \
  --image-url https://... \
  --duration-seconds 8 \
  [--aspect-ratio 16:9] \
  [--wait]

Notes:

Grok duration: 6 or 15
Seedance duration: 5, 8, 10, 12
Seedance in MCP has audio fixed to on

Check Status

python scripts/status.py \
  --request-id YOUR_REQUEST_ID \
  [--download] \
  [--output /tmp/video.mp4]

Idempotency

idempotency_key prevents duplicate billing on retries. If omitted, scripts auto-generate one UUID-based key.

References

安全使用建议

Before installing, note that the scripts require a FILTRIX_MCP_API_KEY (the registry metadata incorrectly omits this). The skill will send any provided image files (base64) and prompts to the Filtrix MCP endpoint and will download video bytes from URLs returned by that service; only supply non-sensitive images or content you are willing to transmit. Verify the MCP endpoint (https://mcp.filtrix.ai/mcp) and that the API key you provide has the minimal necessary scope/credits. If you don't recognize the owner, ask for clearer metadata (declare required env vars) or review the included Python files yourself. If you need a higher assurance level, request the publisher to correct the registry metadata and supply an official homepage or source repository before use.

功能分析

Type: OpenClaw Skill Name: filtrix-video-gen Version: 0.1.0 The filtrix-video-gen skill bundle is a legitimate tool for interacting with the Filtrix Remote MCP service to generate and download videos. The Python scripts (scripts/generate.py, scripts/status.py, and scripts/mcp_client.py) use standard libraries to perform JSON-RPC calls and download media from the authorized endpoint (https://mcp.filtrix.ai/mcp). No evidence of data exfiltration, malicious execution, or prompt injection was found.

能力评估

⚠ Purpose & Capability

Name/description, SKILL.md, and the scripts consistently target Filtrix MCP (https://mcp.filtrix.ai/mcp) and perform text→video, image→video submission, polling, and downloading — that capability is coherent. However the registry metadata claims 'Required env vars: none' and 'Primary credential: none' while both SKILL.md and the code require FILTRIX_MCP_API_KEY (with MCP_API_KEY as a fallback). This mismatch between declared registry requirements and actual runtime requirements is unexpected and should be corrected/clarified.

✓ Instruction Scope

SKILL.md and scripts limit behavior to: call the MCP endpoint, optionally read a local image file (for image-to-video), base64-encode and upload that image, poll get_video_status, and download the video bytes from the returned video URL. The instructions do not reference unrelated system files or extra credentials. The scripts write output files (default under /tmp) and will download whatever URL the MCP returns (expected behavior for this tool).

✓ Install Mechanism

No install spec (instruction-only) and included Python scripts only — nothing is downloaded from external arbitrary URLs at install time. This is lower risk than an installer that pulls remote archives. The skill runs local Python code, which is normal for a script-based integration.

⚠ Credentials

Runtime requires FILTRIX_MCP_API_KEY (and will accept MCP_API_KEY) to authenticate to the Filtrix MCP endpoint; that is proportionate to the stated purpose. The concern is the registry metadata failing to declare this required credential. Also note the scripts will read local image files and upload their base64 content to the remote service — that is expected but users should be aware they may transmit any provided files to Filtrix.

✓ Persistence & Privilege

The skill does not request permanent/always-on inclusion, does not modify other skills or agent-wide settings, and does not persist secrets itself. Autonomous invocation is allowed (platform default) but not combined with other high-risk privileges here.

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 filtrix-video-gen
安装完成后，直接呼叫该 Skill 的名称或使用 /filtrix-video-gen 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

v0.1.0

Initial release of filtrix-video-gen skill. - Enables video generation via Filtrix Remote MCP API. - Supports text-to-video and image-to-video modes. - Includes tools for polling video task status and downloading completed videos. - Requires MCP API key for authentication. - Provides CLI examples and guides for typical workflows.

元数据

Slug filtrix-video-gen

版本 0.1.0

许可证 MIT-0

累计安装 1

当前安装数 1

历史版本数 1

常见问题