← 返回 Skills 市场
视频反向提示
作者
sctc888-hub
· GitHub ↗
· v1.0.0
· MIT-0
115
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install reverseprompt
功能描述
Analyze videos to extract reverse prompts, shot-by-shot breakdowns, and AI-ready visual descriptions via the NanoPhoto.AI Video Reverse Prompt API. Use when:...
安全使用建议
This skill appears coherent and limited to uploading .mp4 files to NanoPhoto.AI using your NANOPHOTO_API_KEY. Before installing: (1) Provide the API key only via the platform's secure env-variable setting (do not paste into chat). (2) Only upload videos you are authorized to share — uploaded content will leave your environment and go to nanophoto.ai. (3) Check the service's pricing/credits and rate limits. (4) Note minor metadata/provenance mismatches in the package (ownerId/slug/version differences); if provenance matters to you, verify the publisher on https://nanophoto.ai and confirm this package was published by the official NanoPhoto vendor. (5) Use the bundled script for local uploads as recommended (it enforces the 30 MB limit and avoids large inline base64 in shells).
功能分析
Type: OpenClaw Skill
Name: reverseprompt
Version: 1.0.0
The skill provides a legitimate interface for the NanoPhoto.AI Video Reverse Prompt API, allowing users to extract shot breakdowns and visual descriptions from videos. The bundled Python script (scripts/reverse_prompt_file.py) correctly implements the documented API workflow, including file size validation, base64 encoding, and retry logic, without any evidence of malicious intent or unauthorized data exfiltration.
能力评估
Purpose & Capability
Name, description, and runtime artifacts align: the skill calls https://nanophoto.ai/api/sora-2/reverse-prompt and requires a single NANOPHOTO_API_KEY which is appropriate for this API-based video analysis purpose. Note: repository metadata (_meta.json) and registry metadata show minor mismatches (ownerId, slug, and version differences) which affect provenance but do not change runtime behavior.
Instruction Scope
SKILL.md instructions are scoped to collecting a video source (YouTube/URL/local .mp4), optionally base64-encoding a local file, and POSTing it to the NanoPhoto API. The instructions explicitly require user authorization for uploads and do not instruct the agent to read unrelated files or environment variables.
Install Mechanism
No install spec (instruction-only) and a small included Python uploader script. The script uses stdlib urllib, validates file extension/size, base64-encodes, and POSTs to the documented endpoint — low install risk.
Credentials
Only NANOPHOTO_API_KEY is required and declared as the primary credential. That is proportionate to the skill's function; no other SECRET/TOKEN/PASSWORD env vars are requested.
Persistence & Privilege
Skill does not request always:true, does not alter other skills' configs, and is not granted elevated persistence. Default autonomous invocation remains (platform normal behavior) but is not combined with other red flags.
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 reverseprompt - 安装完成后，直接呼叫该 Skill 的名称或使用
/reverseprompt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
v1.0.0
Initial release of the Video Reverse Prompt skill.
- Analyze videos to extract reverse prompts, shot-by-shot breakdowns, and AI-ready visual descriptions using the NanoPhoto.AI API.
- Supports video sources from YouTube links, direct .mp4 URLs, and local .mp4 files (≤30 MB).
- Requires configuration of the NANOPHOTO_API_KEY environment variable.
- Provides bundled script for reliable local file uploads and example API calls for all modes.
- Returns streaming Markdown tables summarizing each shot and overall video details.
- Detailed error codes and troubleshooting guidance included.
元数据
常见问题
视频反向提示 是什么？
Analyze videos to extract reverse prompts, shot-by-shot breakdowns, and AI-ready visual descriptions via the NanoPhoto.AI Video Reverse Prompt API. Use when:... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 115 次。
如何安装 视频反向提示？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install reverseprompt」即可一键安装，无需额外配置。
视频反向提示 是免费的吗？
是的，视频反向提示 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
视频反向提示 支持哪些平台？
视频反向提示 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 视频反向提示？
由 sctc888-hub（@sctc888-hub）开发并维护，当前版本 v1.0.0。
推荐 Skills