Skillv1.0.0

ClawScan security

Life Analyzer · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 6, 2026, 9:15 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's declared purpose (analyzing journal entries for tone, patterns, root issues, and action steps) aligns with its instructions and it requests no installs, binaries, or credentials — however it will process very sensitive personal text so exercise privacy and safety caution before use.
Guidance: This skill appears internally coherent, but it will process sensitive personal writing. Before installing: 1) Confirm how the platform handles and stores user inputs and logs (are entries sent to external servers, retained, or logged?). 2) Don’t submit crisis content or suicidal ideation to this skill — it has no crisis-handling instructions; use emergency/professional resources instead. 3) Because the publisher is unknown and there is no homepage, prefer testing with non-sensitive sample text first. 4) If you plan to use it regularly, check the platform’s retention and deletion policies and consider whether you need an audited skill from a known publisher. 5) Consider adding explicit safety constraints (ask user about consent to analyze sensitive material; include a warning/disclaimer) before relying on its output for important decisions.

Purpose & Capability: okName/description match the SKILL.md instructions (detect tone, identify patterns/blind spots, extract root issue, produce actionable steps). No unrelated env vars, binaries, or installs are requested.
Instruction Scope: noteInstructions are narrowly scoped to the provided text and do not instruct reading files, env vars, or contacting external endpoints, but they are vague about safety boundaries (e.g., no guidance for crisis language) and include subjective directives ('extract root problem', 'be honest, not comforting') that could produce blunt or harmful feedback.
Install Mechanism: okNo install spec, no code files — instruction-only. That minimizes disk-level risk; nothing is downloaded or written by the skill itself.
Credentials: okSkill requests no environment variables, credentials, or configuration paths. The requested capabilities are proportional to the stated task.
Persistence & Privilege: okSkill is not marked always:true and uses normal agent invocation. It does not request persistent privileges or modify other skills' configs.