秦丝旺剪 - AI智能视频剪辑

秦丝智能视频剪辑APP的API集成，用于创建和管理AI视频剪辑任务。 TRIGGER when: 用户请求创建视频、生成视频剪辑、查看视频任务列表、下载剪辑结果、等待任务完成、配置旺剪账号。 触发词: "创建视频"、"视频剪辑"、"生成视频"、"查看任务"、"下载视频"、"等待视频"、"配置旺剪"、"旺剪账号"、...

Security Scan

The skill's code and instructions match its stated purpose (integrating with the QinSilk/Wangcut video API); it stores and uses account credentials via a local config file but does not request unrelated credentials or perform unexpected network I/O.

Purpose & Capability

Name/description describe an API client for QinSilk/Wangcut video editing; included Python client (wangcut_api.py), example config, and usage instructions align with that purpose. The default API base_url (cloud.qinsilk.com) matches the description.

Instruction Scope

Runtime instructions tell the agent to import the provided Python module and to create/update a local config.ini with account credentials. The skill reads/writes config.ini in the working directory or skill/project locations (including a project-root lookup), and downloads videos to a local downloads/ directory — all expected for this kind of client but it does mean user credentials and downloaded video files are stored locally.

Install Mechanism

No install spec; this is an instruction+script skill. docs recommend installing the single dependency 'requests' which is reasonable and expected.

Credentials

The skill requests no environment variables or external credentials beyond the QinSilk account provided by the user. It stores the username/password in a local config.ini (written plaintext in setup_config). That storage and the prompt for credentials are proportionate to the stated functionality but are sensitive — the skill does not declare any other credential access.

Persistence & Privilege

always:false and default autonomous invocation are set (normal). The skill writes/updates its own config.ini and creates a downloads/ directory — expected for a client but this is local persistence of credentials and files that the user should manage (e.g., .gitignore).

Assessment

This skill appears to do what it claims (control QinSilk/Wangcut video tasks). Before installing: (1) be aware it asks you to provide your QinSilk account (username/password) and will write them to a local config.ini in your workspace — do not commit that file to source control; add it to .gitignore. (2) The script stores the password in the config file (and MD5-hashes it only when sending to the server), so treat the config file as sensitive. (3) Verify the base_url (defaults to https://cloud.qinsilk.com/aicut/api/v1) is correct for your account and network; if you have concerns, inspect the full wangcut_api.py for any hidden endpoints before use. (4) Install the 'requests' dependency in an isolated environment (venv) and avoid using high-value or shared credentials unless you trust the service. If you want greater safety, use an account with limited permissions or provision temporary credentials.

秦丝智能视频剪辑

通过API操作秦丝智能视频剪辑APP。

首次使用

首次使用会自动检测配置，如果未配置账号密码会提示：

⚠️ 旺剪配置异常: 账号密码未配置
请提供账号密码进行配置，格式：账号 158xxx 密码 xxx

配置账号密码

用户说："配置旺剪账号，账号 158xxx 密码 xxx"

import sys
sys.path.insert(0, '.claude/skills/wangcut/scripts')
from wangcut_api import setup_config

# 配置账号密码（自动创建或更新 config.ini）
config_path = setup_config(
    username="15812345678",
    password="your_password"
)
print(f"配置已保存到: {config_path}")

检查配置状态

import sys
sys.path.insert(0, '.claude/skills/wangcut/scripts')
from wangcut_api import check_config, get_config_status_message

status, message, path = check_config()
print(f"状态: {status}, 消息: {message}")

# 或获取友好提示
print(get_config_status_message())

核心功能

创建视频任务

自动从最近100个素材中随机选择7个：

import sys
sys.path.insert(0, '.claude/skills/wangcut/scripts')
from wangcut_api import create_video_task

task_id = create_video_task("你的视频文案内容")
print(f"任务ID: {task_id}")

查看任务列表

import sys
sys.path.insert(0, '.claude/skills/wangcut/scripts')
from wangcut_api import list_recent_tasks

tasks = list_recent_tasks(count=10, with_details=True)
status_map = {0: '待处理', 1: '处理中', 2: '处理中', 3: '处理中', 4: '已完成', 5: '失败'}
for t in tasks:
    print(f"ID:{t['id']} 状态:{status_map.get(t.get('status'),'未知')}")
    print(f"   文案:{(t.get('script_content') or '')[:30]}...")
    print(f"   时长:{t.get('duration')}秒 分辨率:{t.get('resolution')}")
    print(f"   封面:{t.get('cover_title')} 语音:{t.get('voice_name')}")
    print(f"   播放地址:{t.get('play_url')}")
    print(f"   下载地址:{t.get('video_url')}")

等待完成并下载

import sys
sys.path.insert(0, '.claude/skills/wangcut/scripts')
from wangcut_api import wait_and_download

filepath = wait_and_download(task_id=12575, timeout=600, save_dir="./downloads")
print(f"视频已下载: {filepath}")

任务状态码

码	状态
0	待处理
1-3	处理中
4	已完成
5	失败

高级用法

指定素材ID

import sys
sys.path.insert(0, '.claude/skills/wangcut/scripts')
from wangcut_api import WangcutAPI

api = WangcutAPI()
result = api.create_task(
    script_content="视频文案",
    material_ids=[54131, 54130, 54129],
    voice_speed=1.5,
    subtitle_font_color="white"
)

查看任务详情

import sys
sys.path.insert(0, '.claude/skills/wangcut/scripts')
from wangcut_api import WangcutAPI

api = WangcutAPI()
detail = api.get_task_detail(task_id=12575)
info = api.extract_task_info(detail)
print(f"文案: {info['script_content'][:50]}...")
print(f"时长: {info['duration']}秒, 分辨率: {info['resolution']}")
print(f"封面: {info['cover_title']}, 语音: {info['voice_name']}")
print(f"播放地址: {info['play_url']}")  # 用于浏览器预览
print(f"下载地址: {info['video_url']}")  # 用于下载到本地