Nate B Jones Digest
ReviewAudited by ClawScan on May 1, 2026.
Overview
The skill is a coherent YouTube digest playbook, but it can send digests through configured email/chat/Docs accounts and can be scheduled, so users should review recipients, credentials, and cron settings.
Before running this skill, edit config.yml to use your intended email recipients, chat targets, and document sharing settings; confirm any Google/YouTube credentials and scopes; install external tools only from trusted sources; and enable cron automation only if you want digests sent automatically.
Findings (5)
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.
If run with the provided or customized config, the agent may send digest content or share documents to the configured recipients.
The skill can send email and create/share Google Docs as part of publishing the digest. This is disclosed and aligned with the stated purpose, but it performs outbound actions using configured accounts.
gog gmail send --to "{{config.outputs.email.to}}" ... gog docs share <docId> --email {{config.outputs.doc.share_with}}Review config.yml before use, verify all email/chat/doc recipients, and require confirmation for the first run or any new output channel.
The skill may use the user's YouTube API quota and Google account permissions to poll videos, send mail, create docs, or share docs.
The workflow expects YouTube and Google account access, while the registry metadata declares no required env vars or primary credential. The credential use is purpose-aligned but should be explicitly reviewed.
API credentials: YouTube Data API key (for upload polling), Gmail/Google Docs auth handled via `gog` skill.
Use least-privilege credentials, confirm gog/Gmail/Docs scopes, avoid storing secrets in plain config files, and revoke access if the digest is no longer needed.
A configured cron job could automatically send future digests without asking each time.
The skill supports scheduled recurring operation. This is expected for a monitoring digest, but it means the workflow may continue running after initial setup.
Create a cron job or OpenClaw cron entry using `poll_cron` from config. Each run should: ... publish, log the video ID.
Enable scheduling only intentionally, document where the cron entry lives, and review or disable it if recipients, credentials, or desired behavior change.
Users may need to install or rely on external tools whose versions and provenance are not fixed by this skill.
The instruction-only skill references external tools and libraries but provides no pinned install spec. That is reasonable for a playbook, but users must choose and trust those dependencies themselves.
References ... `youtube_transcript_api` docs ... Whisper CLI ... yt-dlp ... gog
Install dependencies from official sources, pin versions where practical, and review any helper skill such as gog separately.
Local logs may accumulate transcript and summary data over time.
The workflow intentionally persists transcripts, summaries, and last-processed video state. For this public YouTube channel that is proportionate, but it creates retained local context.
Save the raw transcript alongside metadata ... Keep it in your logs for traceability but do not distribute it by default.
Keep logs in a known directory, set a retention policy if needed, and avoid reusing this workflow for private media without revisiting storage and sharing rules.