← 返回 Skills 市场
101
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install wechat-channel-live-replay
功能描述
(已验证) 根据视频号名称搜索用户并下载指定日期的直播回放视频,支持自动转写直播稿。
安全使用建议
This skill appears to implement the advertised functionality, but there are important inconsistencies and privacy implications you should consider before installing:
- The registry entry does not declare the TikHub API token or config path, but SKILL.md and the script require you to store tikhub_api_token in ~/.openclaw/config.json (or provide TIKHUB_TOKEN). If that config file contains other secrets, the script will read it to find the token — review that file first.
- The script accesses the network (api.tikhub.io and media URLs returned by the API) and will download media to your machine and write transcriptions into ~/.openclaw/skills/live-replay-analyzer/input and workspace/data. Ensure you are comfortable with those write locations and have sufficient disk space.
- It requires ffmpeg on PATH and the Whisper package (local transcription). Confirm you install those from trusted sources.
- The SKILL.md source URL is a placeholder; provenance is weak. If you don't fully trust this package, run it in a sandbox or inspect the full export.py yourself. Check the full remainder of the script (the provided file was truncated in the listing) for any unexpected network calls or data exfiltration paths.
Actionable steps before installing:
1) Open and inspect ~/.openclaw/config.json to ensure it only contains keys you are willing to expose to the skill; consider creating a dedicated TikHub token with limited scope.
2) Review the complete export.py file locally for any additional endpoints or behaviors not covered in SKILL.md.
3) Run the tool in an isolated environment (container or VM) if you are unsure, and monitor network activity during first runs.
4) If you rely on a centralized skill registry, ask the publisher to update registry metadata to declare required env vars and config paths so the requirements and permissions are explicit.
功能分析
Type: OpenClaw Skill
Name: wechat-channel-live-replay
Version: 1.0.0
The skill is a legitimate tool designed to download and transcribe WeChat Channel live replays using the TikHub API (api.tikhub.io). The code in export.py correctly handles API authentication via a user-provided token, implements robust input sanitization for filenames to prevent injection during subprocess calls to ffmpeg, and follows the stated purpose in SKILL.md without any evidence of malicious intent or unauthorized data exfiltration.
能力评估
Purpose & Capability
The skill's description (search TikHub, download replays, transcribe) matches the code behavior. However the registry metadata declares no required env vars/config paths/binaries, while SKILL.md and export.py require a TikHub API token, ffmpeg, and Whisper (pip package). The skill also declares a dependency on another skill (live-replay-analyzer / douyin-video-transcriber) in SKILL.md that is not reflected in registry requirements. The mismatch between declared registry requirements and actual runtime needs is an incoherence.
Instruction Scope
SKILL.md and export.py instruct the agent to locate ~/.openclaw/config.json (or a project .openclaw dir), read a tikhub_api_token, call TikHub API endpoints (api.tikhub.io) to search users and fetch live history, download media URLs returned by the API, extract audio with ffmpeg, and save transcriptions into the live-replay-analyzer skills input directory. These actions are in-scope for the stated purpose, but they involve reading a config file and writing into ~/.openclaw/skills/* and workspace/data, and they perform network requests to TikHub and arbitrary media URLs present in the API response — all of which the user should expect and consent to.
Install Mechanism
There is no install spec (instruction-only + included script). No downloads from untrusted URLs or archive extraction are present. Runtime requires pip packages (whisper, requests) and the ffmpeg executable; those are typical for this task. SKILL.md's listed source URL is a placeholder (github.com/your-repo) which reduces provenance confidence.
Credentials
The code requires a sensitive credential (tikhub_api_token) stored in ~/.openclaw/config.json or the TIKHUB_TOKEN env var but the registry metadata lists no required env variables or config paths — this is a clear omission. The script reads the config.json file (and will search parent directories), so it may have access to other keys stored there; while the code only uses tikhub_api_token, reading config files is sensitive and should be disclosed. No other credentials are requested, which is proportionate to the function, but the missing declaration is the issue.
Persistence & Privilege
The skill does not request 'always: true' and does not modify other skills' configuration. It writes downloaded videos and transcriptions into ~/.openclaw/workspace/data and ~/.openclaw/skills/live-replay-analyzer/input, which is expected for integration but is a write access to the agent's skill workspace that the user should be aware of. Autonomous invocation is enabled by default (normal) — combine that with network access and token use when considering risk.
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 wechat-channel-live-replay - 安装完成后,直接呼叫该 Skill 的名称或使用
/wechat-channel-live-replay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
wechat-channel-live-replay v1.0.0
- 首次发布:支持根据微信视频号名称搜索用户并下载指定日期的直播回放视频。
- 支持自动转写直播视频为文字稿,输出到指定目录,便于后续复盘分析。
- 可通过参数选择是否转写或仅下载视频文件。
- 输出包含:回放视频、文字稿、和详细的汇总报告(JSON)。
- 与 live-replay-analyzer 技能深度集成,实现自动化复盘数据流。
元数据
常见问题
WeChat Channel Live Replay 是什么?
(已验证) 根据视频号名称搜索用户并下载指定日期的直播回放视频,支持自动转写直播稿。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 101 次。
如何安装 WeChat Channel Live Replay?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install wechat-channel-live-replay」即可一键安装,无需额外配置。
WeChat Channel Live Replay 是免费的吗?
是的,WeChat Channel Live Replay 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
WeChat Channel Live Replay 支持哪些平台?
WeChat Channel Live Replay 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 WeChat Channel Live Replay?
由 an(@ahsbnb)开发并维护,当前版本 v1.0.0。
推荐 Skills