← 返回 Skills 市场
Podcast Intel
作者
claw_for_le_use_nex
· GitHub ↗
· v1.0.0
· MIT-0
257
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install podcast-intel
功能描述
Automation skill for Podcast Intel.
安全使用建议
This skill appears to do what it says: fetch RSS feeds, download audio, transcribe with Whisper/OpenAI, analyze content, and store a local diary. Before installing or running: 1) Be prepared to provide an OPENAI_API_KEY (the skill sends audio/text to OpenAI when using the API). 2) The included install.sh will pip-install dependencies and may attempt to install ffmpeg via your system package manager (may require sudo on Linux/macOS and a working network). 3) The skill stores caches, transcripts, and diary files under OPENCLAW_HOME (default ~/.openclaw) — review the example config files (config/*.example.yaml) and set PODCAST_INTEL_FEEDS / PODCAST_INTEL_INTERESTS if you want custom locations. 4) If you are concerned about sending audio to OpenAI, use the WHISPER_USE_LOCAL option and install a local whisper model, but note local transcription has its own resource/installation needs. 5) As with any third-party code, inspect the scripts locally before running install.sh and avoid running scripts as root unless you understand the commands.
功能分析
Type: OpenClaw Skill
Name: podcast-intel
Version: 1.0.0
The podcast-intel skill is a well-structured tool for fetching, transcribing, and analyzing podcast episodes. It demonstrates good security practices, such as implementing SSRF protection in `scripts/transcribe.py` to validate audio URLs before processing them with ffmpeg. The installation and orchestration logic in `install.sh` and `scripts/main.py` are transparent, using standard system package managers and safe subprocess execution. No evidence of data exfiltration, malicious persistence, or prompt injection was found.
能力评估
Purpose & Capability
Name/description (podcast transcription, segmentation, scoring, diary) match the included scripts. Required env var (OPENAI_API_KEY) and binaries (python3, ffmpeg) are appropriate for using Whisper/OpenAI and ffmpeg-based audio fetching/processing. Optional env vars referenced in code (OPENAI_BASE_URL, WHISPER_MODEL, WHISPER_USE_LOCAL, OPENAI_MODEL) are consistent with the stated purpose.
Instruction Scope
SKILL.md tools map to local scripts (scripts/main.py, diary.py, etc.). The runtime instructions operate on RSS feeds, download audio, call Whisper/OpenAI for transcription, analyze text, and append to a local diary under the OpenClaw directory — all within the scope of a podcast intelligence engine. The code validates audio URLs to block local/private IPs before ffmpeg fetches, and there are no instructions to read unrelated system files or unrelated environment secrets.
Install Mechanism
This is an instruction-only skill in the registry metadata, but the bundle includes an install.sh that installs Python dependencies (pip install -r requirements.txt) and attempts to auto-install ffmpeg via system package managers (brew/apt-get/dnf). The install script may require elevated privileges for OS package installs (sudo) and downloads packages from PyPI; those operations are expected for this functionality but are outside the registry's automated install spec and will run on the user's machine if they run install.sh.
Credentials
Only OPENAI_API_KEY is declared as required and is actually used (OpenAI API calls). Other env vars referenced are optional and justified (endpoint override, whisper model selection, local whisper toggle). No unrelated credentials or broad secrets are requested.
Persistence & Privilege
The skill does not request always:true and does not modify other skills. It creates and writes to its own directories under OPENCLAW_HOME (default ~/.openclaw) for cache, workspace, memory, and diary — this is normal for a local data pipeline and documented in install/setup scripts.
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 podcast-intel - 安装完成后，直接呼叫该 Skill 的名称或使用
/podcast-intel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
podcast-intel 1.0.0 - Initial Release
- Launches a podcast intelligence engine for transcribing, segmenting, summarizing, and scoring podcast episodes from RSS feeds.
- Provides ranked "worth your time" podcast recommendations with cross-source deduplication.
- Maintains a local podcast consumption diary with structured logs and topics.
- Supports episode analysis by URL and instant recommendations from cached transcripts.
- Multiple output formats available: json, markdown, and tts.
元数据
常见问题
Podcast Intel 是什么？
Automation skill for Podcast Intel. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 257 次。
如何安装 Podcast Intel？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install podcast-intel」即可一键安装，无需额外配置。
Podcast Intel 是免费的吗？
是的，Podcast Intel 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Podcast Intel 支持哪些平台？
Podcast Intel 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 Podcast Intel？
由 claw_for_le_use_nex（@nexspaceiosdev）开发并维护，当前版本 v1.0.0。
推荐 Skills