Skillv1.0.2

ClawScan security

Veteran Entrepreneur Video · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousApr 8, 2026, 2:52 PM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's stated purpose (AI video creation for financial advisors) is plausible, but the instruction-only bundle declares no dependencies, no credentials, and no install steps — which is unusual for a video/TTS/generation workflow and creates unanswered questions about how it will actually produce videos and whether it will handle sensitive client data.
Guidance: This skill claims to create marketing videos but declares no integrations, credentials, or installers — ask the developer (or inspect the full SKILL.md) how videos are actually produced: which services/APIs or binaries it uses, what credentials (if any) you'll need, and whether the skill will ever request or store client financial data. If it will call third-party APIs, require: 1) a list of endpoints and the exact env vars/keys needed, 2) privacy/compliance assurances for handling PII/financial info, and 3) a sample run or transcript showing what data the agent requests. Do not provide real client account numbers, Social Security numbers, or full financial records until you confirm where data is sent and how it is stored/retained.

Review Dimensions

Purpose & Capability: noteName and description match a legitimate use: creating marketing/educational videos for financial advisors. However, producing videos (TTS, stock footage, video rendering, hosting) normally requires external services, binaries, or credentials; none are declared. This mismatch is noteworthy but could be explained if the skill only generates scripts/templates rather than actual video files.
Instruction Scope: noteThe provided SKILL.md text is heavily descriptive and use-case focused. Because this is an instruction-only skill with no code files, the runtime instructions (not fully shown) are the entire attack surface. The absence of declared env vars or endpoints means either the instructions: (a) generate text/scripts only, (b) expect platform-provided tools (not declared), or (c) omitted needed integration details. If the instructions ask the agent to solicit or embed client-specific financial data (PII/financial accounts), that raises privacy and compliance concerns; that behavior was not visible in the truncated content.
Install Mechanism: okNo install spec and no code files — lowest install risk. Nothing will be written to disk by an installer because there is no install step declared.
Credentials: concernZero required environment variables or credentials are declared. For genuine video creation workflows you'd expect API keys (TTS, video render, asset storage/CDN, stock footage) or at least notes explaining reliance on platform services. The lack of declared credentials could mean the skill is limited to producing scripts only, or it fails to declare access it actually needs — either way this is disproportionate to the advertised capability.
Persistence & Privilege: okSkill is not always-enabled and does not request elevated persistence. Autonomous invocation is allowed (platform default) but not combined with other high-risk elements here. No indicators it modifies other skills or system settings.