ln-722-backend-generator
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
dotnet buildduring the verification phase to ensure the generated solution compiles correctly. This is standard behavior for a development tool. - [EXTERNAL_DOWNLOADS]: The skill references NuGet packages from
nuget.org. As this is the official .NET package registry, it is considered a well-known and trusted service. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes untrusted entity names and project configurations. These inputs are used to generate file names and source code content. Although the skill documentation includes a risk mitigation step to validate entity names, there is a potential for malicious strings to be embedded in the generated output if sanitization is bypassed. Documentation of evidence chain: (1) Ingestion:
projectName,entities, andtargetPathin SKILL.md. (2) Boundaries: None specified in prompt instructions. (3) Capabilities: File writing anddotnet buildexecution. (4) Sanitization: Mentioned as a risk mitigation step but not explicitly implemented in the provided patterns.
Audit Metadata