← 返回 Skills 市场
3419
总下载
5
收藏
1
当前安装
17
版本数
在 OpenClaw 中安装
/install youtube-apify-transcript
功能描述
Fetch YouTube transcripts via APIFY API. Works from cloud IPs (Hetzner, AWS, etc.) by bypassing YouTube's bot detection. Features local caching (FREE repeat...
安全使用建议
This skill appears to be internally consistent and implements exactly what it claims: it needs an APIFY API token and the Python requests package to call Apify actor(s) and cache results. Before installing:
- Provide an APIFY_API_TOKEN (SKILL.md and package.json require it); ignore the registry summary that says none — that is inconsistent.
- Review or set YT_TRANSCRIPT_CACHE_DIR if you don't want transcripts stored under the skill directory; cached transcripts persist until you run --clear-cache.
- Understand that transcripts are retrieved via a third-party Apify actor (actor IDs are listed). Using Apify means your requests go through Apify's infrastructure and the actor code run by that actor owner; ensure you trust that actor/provider and are comfortable with any data sent to Apify and potential billing on your account.
- The tool does not request other credentials or perform other surprising actions in the visible code, but you may want to inspect the remainder of the script (truncated in the provided snippet) to confirm there are no unexpected network endpoints or behavior before granting the APIFY token.
If you need stronger guarantees, run the script locally with a limited Apify token or instrument the network calls to confirm only api.apify.com is contacted.
功能分析
Type: OpenClaw Skill
Name: youtube-apify-transcript
Version: 1.3.3
The skill bundle is a legitimate tool for fetching YouTube transcripts via the Apify API. The Python script (scripts/fetch_transcript.py) implements standard API interaction, local caching, and batch processing without any signs of malicious intent, obfuscation, or unauthorized data access. It correctly handles the APIFY_API_TOKEN via environment variables and communicates only with official Apify endpoints (api.apify.com).
能力评估
Purpose & Capability
The skill's declared functionality (fetch YouTube transcripts via Apify) matches the code and instructions: it calls the Apify API and caches results. One inconsistency: the top-level registry summary in the provided metadata listed 'Required env vars: none', while SKILL.md and package.json correctly declare APIFY_API_TOKEN (required) and an optional YT_TRANSCRIPT_CACHE_DIR. This appears to be a metadata oversight rather than functional misalignment.
Instruction Scope
SKILL.md instructs the agent to run the included Python script, install the requests library, and set APIFY_API_TOKEN. The script (visible portion) only reads/writes local cache files in the skill directory and talks to the Apify API. Agent instructions explicitly say to fetch transcripts with the script then summarize locally. No instructions request unrelated system files, other credentials, or external endpoints beyond Apify.
Install Mechanism
There is no remote download/install step in the registry install spec. The only required binary is python3 and the only Python dependency is 'requests' (pip install requests), which is proportionate and low-risk. The skill ships a local script rather than downloading code from arbitrary URLs.
Credentials
The single required secret is APIFY_API_TOKEN, which is appropriate for calling the Apify API. The optional YT_TRANSCRIPT_CACHE_DIR is reasonable. However, the registry summary conflictingly claimed no required env vars — users should rely on SKILL.md/package.json and provide APIFY_API_TOKEN. No other unrelated secrets or credentials are requested.
Persistence & Privilege
The skill persists cached transcripts as JSON files under the skill directory (.cache/ by default) or under a directory set by YT_TRANSCRIPT_CACHE_DIR. This is normal for a caching tool but means fetched transcript contents will remain on disk unless cleared with --clear-cache; users should consider where the skill is installed and who can access that directory. always is false and the skill does not request elevated agent privileges.
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 youtube-apify-transcript - 安装完成后，直接呼叫该 Skill 的名称或使用
/youtube-apify-transcript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
v1.3.3
- Added explicit requirement and install instructions for the Python requests package.
- Updated description and metadata to clarify requests dependency.
- Improved setup documentation for easier installation.
- Bumped version to 1.3.3.
v1.3.2
- Removes support and documentation for AI-powered transcript summarization features and the --summarize flag.
- Updates usage examples and instructions to reflect the removal of summarization options.
- Removes requirement for the "claude" CLI tool (now only requires python3).
- Adds new agent instructions for handling transcript summaries directly.
- Updates metadata and environment requirements accordingly.
v1.3.1
### youtube-apify-transcript 1.3.1
- Added support and documentation for optional AI summarization using the `claude` CLI.
- Updated metadata to require the `claude` binary for summary features.
- Improved SKILL.md description and tags for more accurate feature listing.
v1.3.0
**YouTube transcript fetcher now supports on-the-fly transcript summarization with flexible styles and caching.**
- Added `--summarize` option to generate Claude summaries of transcripts (brief, detailed, bullets, or tldr; custom language and model supported).
- Summary data is cached alongside transcripts to avoid repeat AI costs.
- All summary options are documented in usage instructions, including new CLI flags.
- JSON output now optionally includes summary content.
v1.2.0
Switch default actor to topaz_sharingan~Youtube-Transcript-Scraper-1 for better transcript reliability. Old karamelo actor often returns null captions for new videos. Also adds support for multiple actor input formats.
v1.1.3
Improved auth/cache docs and synced metadata cleanup changes.
v1.1.2
Fix: SKILL.md cache path updated to match code. APIFY_API_TOKEN declared in registry metadata.
v1.1.1
Security: cache dir now skill-local (configurable via env). APIFY token moved from query string to Authorization Bearer header.
v1.1.0
Added local transcript caching (permanent, by video ID), --no-cache, --clear-cache, --cache-stats. New batch mode: --batch file.txt for bulk processing.
v1.0.7
- Added a CHANGELOG.md file.
- Updated SKILL.md to include a Metadata section with structured YAML and an emoji.
- Simplified metadata requirements in SKILL.md (now only lists required environment variable and binary).
- Bumped version to 1.0.7.
v1.0.6
Add ClawHub runtime requirements metadata (SKILL.md frontmatter)
v1.0.5
Add runtime requirements metadata for ClawHub discovery
v1.0.4
Fix: Handle string captions format from APIFY API. The API returns captions as plain strings instead of dicts with timestamps. Updated both text and JSON formatters to handle both formats gracefully.
v1.0.3
youtube-apify-transcript 1.0.3
- No changes detected from previous version 1.0.2.
- Version number updated only; functionality and documentation remain the same.
v1.0.2
- Added a frontmatter block with name, version, description, and tags at the top of SKILL.md.
- No changes to functionality or usage instructions. Documentation only.
v1.0.1
Fixed: Now in own GitHub repo (was incorrectly pushed as PR)
v1.0.0
Initial release - Fetch YouTube transcripts via APIFY API
元数据
常见问题
Youtube Apify Transcript 是什么？
Fetch YouTube transcripts via APIFY API. Works from cloud IPs (Hetzner, AWS, etc.) by bypassing YouTube's bot detection. Features local caching (FREE repeat... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 3419 次。
如何安装 Youtube Apify Transcript？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install youtube-apify-transcript」即可一键安装，无需额外配置。
Youtube Apify Transcript 是免费的吗？
是的，Youtube Apify Transcript 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Youtube Apify Transcript 支持哪些平台？
Youtube Apify Transcript 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 Youtube Apify Transcript？
由 Robby（@robbyczgw-cla）开发并维护，当前版本 v1.3.3。
推荐 Skills