Back to skill
Skillv1.0.0
ClawScan security
Game Design KPI Coverage Audit · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 22, 2026, 8:54 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill is instruction-only and its requirements, included reference files, and runtime instructions match its stated purpose of auditing KPI coverage for game features.
- Guidance
- This is an instruction-only audit tool that uses only the included reference files and does not require credentials or installs. It appears coherent and low-risk, but consider what data you will feed it: if you supply sensitive internal metrics or proprietary roadmaps to the agent, those inputs could be exposed according to your environment's data-handling rules. If you want stricter control, limit the agent's autonomous invocation or avoid pasting sensitive dashboards and raw credentials into prompts.
Review Dimensions
- Purpose & Capability
- okName, description, and deliverables align with the required artifacts and included reference docs; there are no unexpected environment variables, binaries, or external services requested.
- Instruction Scope
- okSKILL.md stays on-topic: it instructs the agent to read the bundled reference files and produce an audit (feature read, KPI framing, blind spots, recommendations). It does not request system-wide file access, credentials, or transmission to external endpoints.
- Install Mechanism
- okNo install spec or code files are included; this is an instruction-only skill so nothing will be written to disk or downloaded during install.
- Credentials
- okThe skill declares no required environment variables, credentials, or config paths; the actions described do not imply a need for secrets or unrelated service access.
- Persistence & Privilege
- okalways is false and model invocation is allowed (platform default). The skill does not request persistent system privileges or modifications to other skills' settings.