Token导航 LogoToken导航TokenDH.com
开发敏感数据clawhub未标认证来源可访问clear审计通过

reversepromptreverseprompt 分析

Agent Skill

用于辅助提示词、系统指令、Agent 行为约束和工作流模板的整理。它适合让 Agent 规范任务边界、统一输出格式、拆分操作步骤或优化提示词可复用性。使用时需要保留真实业务约束,不要把示例当硬规则;涉及自动执行、外部工具或高风险操作时,应在提示词中明确确认步骤、权限边界和失败处理方式。

总安装

3,213

周安装

138

GitHub Stars

公开资料未说明

下载量

1,126
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:reverseprompt(reverseprompt 分析)
来源仓库:https://github.com/sctc888-hub/reverseprompt
安装命令:
openclaw skills install reverseprompt
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install reverseprompt

简介

通过视频分析提取反向提示与视觉描述的AI工具。

  • 适用于视频素材预处理、镜头级优化与AI生成内容准备场景。
  • 输出逐帧分析与故障定位,提升后续生成任务效率。
  • 安装命令:openclaw skills install reverseprompt,需上传视频文件或链接。
  • 注意视频格式兼容性,避免因编码问题导致分析失败或结果失真。

SKILL.md

name
video-reverse-prompt
description
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: (1) User wants reverse prompt, prompt extraction, or video-to-prompt analysis, (2) User provides a YouTube link, direct MP4 URL, or local MP4 file and wants a shot list / storyboard-style breakdown, (3) User mentions video analysis, shot breakdown, extract prompt from video, NanoPhoto, or reverse engineer a video prompt. Supports YouTube links, direct .mp4 URLs, and local file upload. Prerequisite: Obtain an API key at https://nanophoto.ai/settings/apikeys and configure env.NANOPHOTO_API_KEY.
homepage
https://nanophoto.ai
metadata
{"openclaw":{"homepage":"https://nanophoto.ai","requires":{"env":["NANOPHOTO_API_KEY"]},"primaryEnv":"NANOPHOTO_API_KEY"}}

Video Reverse Prompt

Analyze videos to extract detailed shot breakdowns and AI-ready prompts via the NanoPhoto.AI API.

Use the bundled script for local file uploads instead of inlining large base64 payloads directly in the shell; it is more reliable for multi-megabyte videos.

Prerequisites

  1. Obtain an API key at: https://nanophoto.ai/settings/apikeys
  2. Configure NANOPHOTO_API_KEY before using the skill.

Preferred OpenClaw setup:

  • Open the skill settings for this skill
  • Add an environment variable named NANOPHOTO_API_KEY
  • Paste the API key as its value

Equivalent config shape:

{
  "skills": {
    "entries": {
      "video-reverse-prompt": {
        "enabled": true,
        "env": {
          "NANOPHOTO_API_KEY": "your_api_key_here"
        }
      }
    }
  }
}

Other valid ways to provide the key:

  • Shell: export NANOPHOTO_API_KEY="your_api_key_here"
  • Tool-specific env config: any runtime that injects NANOPHOTO_API_KEY

Credential declaration summary:

  • Required env var: NANOPHOTO_API_KEY
  • Primary credential: NANOPHOTO_API_KEY
  • No unrelated credentials are required

If the env var is not set, ask the user to configure it before proceeding.

Workflow

  1. Collect the video source from the user (YouTube link, direct .mp4 URL, or local file path)
  2. Determine videoSource type: youtube, url, or file
  3. Confirm the user is authorized to upload or process the content
  4. If local file: read and base64-encode it (must be .mp4, max 30 MB)
  5. Call the API (streaming response)
  6. Return the shot breakdown and prompts to the user

API Call

YouTube Video

curl -X POST "https://nanophoto.ai/api/sora-2/reverse-prompt" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $NANOPHOTO_API_KEY" \
  --data-raw '{
    "videoSource": "youtube",
    "locale": "en",
    "videoUrl": "https://www.youtube.com/watch?v=XXXXXXXXXXX"
  }'

Direct Video URL

curl -X POST "https://nanophoto.ai/api/sora-2/reverse-prompt" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $NANOPHOTO_API_KEY" \
  --data-raw '{
    "videoSource": "url",
    "locale": "en",
    "videoUrl": "https://example.com/video.mp4"
  }'

Local File (Preferred: bundled script)

python3 scripts/reverse_prompt_file.py your-video.mp4 --locale en

The script reads NANOPHOTO_API_KEY from the environment, validates the file size/format, base64-encodes the MP4, and prints the streaming text response.

Local File (Manual Base64 request)

VIDEO_BASE64=$(base64 < your-video.mp4)

curl -X POST "https://nanophoto.ai/api/sora-2/reverse-prompt" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $NANOPHOTO_API_KEY" \
  --data-raw "{
    \"videoSource\": \"file\",
    \"locale\": \"en\",
    \"videoFile\": \"$VIDEO_BASE64\",
    \"videoFileName\": \"your-video.mp4\"
  }"

Parameters

ParameterTypeRequiredDescription
videoSourcestringYesyoutube, url, or file
localestringNoOutput language (default: en). Supported: en, zh, zh-TW, ja, ko, es, fr, de, pt, ru, ar
videoUrlstringConditionalYouTube link or direct .mp4 URL
videoFilestringConditionalBase64-encoded video (when videoSource is file)
videoFileNamestringNoOriginal filename for uploaded videos

Constraints

  • Only .mp4 format supported
  • Max file size: 30 MB (before base64 encoding)
  • videoFile accepts plain base64 or Data URL (data:video/mp4;base64,...)
  • Costs 1 credit per API call

Response

The API returns a streaming text response (Content-Type: text/plain; charset=utf-8) containing a Markdown table with:

  • Shot number, framing/angle, camera movement
  • Detailed visual description
  • Audio analysis (BGM, sound effects, narration)
  • Duration per shot
  • Overall summary

Error Handling

errorCodeHTTPCauseAction
LOGIN_REQUIRED401Invalid or missing API keyVerify key at https://nanophoto.ai/settings/apikeys
API_KEY_RATE_LIMIT_EXCEEDED429Rate limit exceededWait and retry
INSUFFICIENT_CREDITS402Not enough creditsTop up credits
INVALID_INPUT400Missing required parametersCheck videoSource and corresponding fields
INVALID_YOUTUBE_URL400Invalid YouTube URLAsk user for a valid YouTube link
INVALID_VIDEO_URL400Invalid video URLAsk user for a valid .mp4 URL
INVALID_FORMAT400Not MP4 formatOnly .mp4 is supported
FILE_TOO_LARGE400File exceeds 30 MBAsk user for a smaller file
VIDEO_DOWNLOAD_FAILED400Cannot download videoCheck URL accessibility
VIDEO_PROCESSING_FAILED422Processing errorRetry or try a different video
AI_SERVICE_ERROR503AI service unavailableRetry later

Bundled script

  • scripts/reverse_prompt_file.py: Reliable local-file uploader for .mp4 inputs. Use it when the user provides a local video path.

Full API Reference

See references/api.md for complete endpoint documentation.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

85.47%
按下载量换算962

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills