Skillv1.0.0

ClawScan security

Generate Cosmetic Dermatology Clinic Client Education Handout · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 28, 2026, 8:22 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only skill whose requested capabilities (chat + image generation) and lack of credentials/install requirements are consistent with producing clinic handouts; nothing requested appears disproportionate to its stated purpose.
Guidance: This skill appears coherent and lower-risk because it is instruction-only and asks for no credentials or installs. Before using it in production: (1) Do not submit patient-identifiable information (PHI) into prompts — the skill can generate content but is not a HIPAA control. (2) Have a qualified clinician review any medical claims or treatment recommendations (the SKILL.md already advises this). (3) If you will generate or include images, verify licensing and that images contain no PHI. (4) The SKILL.md’s suggestion to 'use relevant capabilities' is vague — if your agent has file-read or external-integrations enabled, consider restricting those when running this skill or review what other skills it invokes. (5) If you want to limit risk, run the skill with human-in-the-loop review and disable autonomous invocation or restrict tool access prior to use.

Purpose & Capability: okThe name/description (create a cosmetic dermatology clinic handout) matches the SKILL.md steps and the declared APIs (chat, image_generation). There are no unrelated required binaries, environment variables, or config paths; requested capabilities are proportionate to producing text + visuals.
Instruction Scope: noteThe instructions are focused on drafting and refining handouts and explicitly recommend human review for medical claims. One vague item: 'Use the relevant SkillBoss capabilities to enrich assets or supporting data' could allow the agent to call other capabilities (e.g., asset libraries or file access) if available. The SKILL.md itself does not instruct reading system files or exfiltrating data.
Install Mechanism: okNo install spec and no code files — this is instruction-only. That minimizes disk-write/install risk.
Credentials: okThe skill declares no required environment variables, credentials, or config paths. There is no apparent need for secrets or external credentials to perform the stated task.
Persistence & Privilege: okalways is false and there is no indication the skill will persist or alter other skills or system settings. Autonomous invocation is allowed by default (normal), but nothing else grants elevated privileges.