Back to skill
Skillv1.0.0
ClawScan security
Customer Feedback Analysis · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 8:26 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This is an instruction-only skill for analyzing user-provided customer feedback, with no code, credentials, install scripts, or hidden capabilities shown.
- Guidance
- This skill appears safe to install based on the provided artifacts. As with any feedback-analysis workflow, avoid pasting unnecessary customer personal information unless it is needed for the task.
Review Dimensions
- Purpose & Capability
- okThe stated purpose and SKILL.md content consistently describe customer review and feedback analysis, sentiment reporting, pain point extraction, and improvement prioritization.
- Instruction Scope
- okThe instructions are limited to accepting user-provided review or business context and producing analysis frameworks and reports; they do not direct the agent to override user intent or perform unrelated actions.
- Install Mechanism
- okThere is no install spec, helper code, package dependency, or executable component beyond the normal ClawHub install command for an instruction-only skill.
- Credentials
- okNo local files, credentials, environment variables, network services, or platform accounts are requested by the artifacts.
- Persistence & Privilege
- okThe artifacts show no persistence, background activity, privileged access, credential handling, or long-running behavior.