data-seeder-generator
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local shell commands such as
find,grep,basename, andlsto identify project structures, namespaces, and existing code patterns. These commands are executed locally to automate project discovery and are scoped to the current working directory. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by ingesting untrusted data from the local codebase to influence its generation logic.
- Ingestion points: The instructions direct the agent to scan and read 'EVERY existing seeder' and 'each entity' file in the repository (e.g., in Step 0.3 and 0.4).
- Boundary markers: There are no markers or instructions that define boundaries between legitimate source code and potentially malicious instructions that could be embedded in the analyzed files.
- Capability inventory: The skill possesses the capability to write new C# files to the filesystem based on patterns it extracts from existing codebase content.
- Sanitization: The skill lacks any defined validation or sanitization process for content read from the repository before it is used in the generation of new code files.
Audit Metadata