Security audit
Security checks for vulnerabilities and agentic risk
ClawScan rated this skill benign with medium confidence: the scanners found no AIG issues and judged its purpose and privilege profile acceptable, while noting an unpinned install command and a possible disclosure gap.
Consider whether you are comfortable using an install command that does not pin the `skills` package version, and avoid including sensitive or identifying character details if they may be sent to an external generation service. The scan was partial: it inspected the two reported Markdown components, but a local reference in SKILL.md could not be resolved, so the evidence is incomplete.
The installation command runs an unpinned skills package through npx, allowing the resolved package version to change over time. A compromised future release could execute code with the installing user's permissions; this is a dependency supply-chain risk, although the finding incorrectly describes the installer as an MCP server.
This step instructs the agent to generate a model sheet using Seedream through the varg workflow, rather than merely listing an available model. Although the file identifies the varg connection, it does not warn users that character details incorporated from their requests will be transmitted to an external generation service.
