Digest
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only digest skill is coherent and not malicious, but it stores learned preferences and may use scheduled updates, so review those settings.
This appears safe to install as an instruction-only personalization skill. Before using it heavily, confirm what preferences it records, where digests should be delivered, and whether scheduled or urgent updates are desired.
Findings (2)
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.
Your reading interests, delivery preferences, and context profiles may be stored and reused in later digests.
The skill maintains persistent preference data that can affect future outputs. This is purpose-aligned for personalization, but users should understand that topics, sources, timing, and channel preferences may be retained.
Auto-evolving user preferences. Edit as you learn.
Review preferences.md periodically, avoid storing sensitive personal or work context unless needed, and keep the confirmation process for important preference changes.
If scheduled delivery is enabled, the agent may send digest updates proactively according to configured preferences.
The skill contemplates scheduled/proactive delivery. This is disclosed and aligned with a digest assistant, but users should keep timing and delivery channels explicit.
Timing per user preference: - Morning digest, evening digest, or both - Weekday vs weekend differences - On-demand vs scheduled
Only enable schedules, urgent alerts, and delivery channels you actually want, especially for group chats, email, or work contexts.