testability-obstacle
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted C# source code and test files from the target workspace to perform analysis and refactoring. This ingestion point creates an attack surface for indirect prompt injection where malicious instructions embedded in the source code (e.g., in comments or string literals) could influence the agent's behavior during refactoring or testing.\n
- Ingestion points: Target production path and existing tests read during Step 1 in
SKILL.md.\n - Boundary markers: The instructions do not specify any delimiters or warnings to ignore embedded instructions in the ingested code.\n
- Capability inventory: The skill modifies production source code, creates new test files, and executes repository-level test commands (subprocess execution).\n
- Sanitization: No explicit sanitization or validation of the untrusted code content is mentioned before it is processed or used in prompts.
Audit Metadata