Podcast Intel
PassAudited by ClawScan on May 10, 2026.
Overview
Podcast Intel appears purpose-aligned, but it uses an OpenAI API key, downloads/transcribes podcast audio, and keeps local diary/cache files.
This skill looks coherent for podcast analysis. Before installing or running it, confirm you are comfortable with it using your OpenAI key, downloading podcast audio, storing local transcripts and diary notes, and optionally installing ffmpeg/Python dependencies if you run install.sh.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Running transcriptions can use your OpenAI account and may incur API costs.
The skill uses the user's OpenAI API key to call the transcription provider, which is expected for this podcast transcription purpose but grants billing/account authority to the skill.
client = OpenAI(api_key=config["api_key"], base_url=config["base_url"])
Use an API key intended for this purpose, monitor usage, and avoid configuring an untrusted OPENAI_BASE_URL.
Podcast feeds can cause the skill to download and process remote audio, consuming bandwidth, time, and provider quota.
The skill invokes ffmpeg on podcast audio URLs. This is central to transcription and uses argument-list subprocess calls plus URL validation, but it still processes remote media selected through feeds or user input.
cmd = [
"ffmpeg",
...
"-i",
url,Use trusted podcast feeds and review configured feed URLs before running large briefings.
If you run the installer, it may install software into your system or Python environment.
The optional installer can modify the system package environment and install Python dependencies. This is normal setup behavior for an audio-processing skill, but it is not represented as a registry install spec.
sudo apt-get install -y ffmpeg ... pip install -r "$SKILL_ROOT/requirements.txt" --quiet
Review install.sh and requirements.txt first, and consider using a virtual environment for Python dependencies.
Your podcast interests and briefing history can remain on disk and influence future recommendations.
The skill persists podcast diary information into OpenClaw memory in addition to a JSONL diary. This supports the recommendation feature but stores listening history and generated topic summaries for later reuse.
# Also write to markdown memory file ... save_markdown_note(markdown)
Use dry_run when you do not want diary writes, and periodically review or clear the podcast-intel cache, workspace, and memory directories if needed.