Back to skill
Skillv1.0.0
ClawScan security
ROI Monitor · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 8:07 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This is an instruction-only marketing ROI analysis skill with no code, credentials, account access, persistence, or hidden execution behavior shown in the artifacts.
- Guidance
- This appears safe to install from the provided artifacts. Treat its budget recommendations as advisory, and avoid pasting confidential business data unless you are comfortable using it in the conversation.
Review Dimensions
- Purpose & Capability
- okThe stated purpose—building omnichannel ROI comparisons and budget recommendations—matches the SKILL.md instructions and reference files.
- Instruction Scope
- okInstructions focus on tables, attribution caveats, guardrails, and user-facing recommendations; they do not attempt to override user intent or force unsafe actions.
- Install Mechanism
- okThere is no install spec, no required binaries, no required environment variables, and no code files to execute.
- Credentials
- okThe skill relies on user-provided marketing and business context, with no artifact evidence of filesystem, network, credential, or account access.
- Persistence & Privilege
- okNo persistence, background operation, credential use, or privilege escalation mechanism is present in the supplied artifacts.