sparkscan-maui
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch the latest package version information from the official NuGet gallery (nuget.org) and retrieve detailed API documentation from Scandit's official documentation site (docs.scandit.com). These network operations target well-known and vendor-specific sources to ensure technical accuracy and are standard practice for development-focused skills.\n- [PROMPT_INJECTION]: The skill includes instructions to prioritize provided reference files over internal training data ("Do Not Trust Internal Knowledge"). This is a safety and accuracy grounding technique designed to prevent the generation of hallucinated or outdated API signatures and is not malicious.\n- [COMMAND_EXECUTION]: The skill guides the agent in modifying project files (.csproj, .xaml, .cs) to integrate the scanning library. These operations, such as adding NuGet references and configuring build properties like SupportedOSPlatformVersion, are within the intended scope of a developer tool and do not involve unauthorized command execution.\n- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests and processes user-provided source code. However, the risk is minimized as the skill focuses on structured library integration tasks.\n
- Ingestion points: User project files including .csproj, .xaml, and .cs source files.\n
- Boundary markers: Not explicitly defined in the prompt instructions to separate user content from system instructions.\n
- Capability inventory: The skill is capable of modifying local files and performing network fetches for documentation via the agent's built-in tools.\n
- Sanitization: No specific sanitization or validation of the ingested code content is performed by the skill instructions.
Audit Metadata