file-organization
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill provides automated file organization capabilities by reading existing directory structures and file contents. This creates a surface for indirect prompt injection, as an attacker could place a malicious file within a project to influence the agent's behavior during a refactoring task.\n
- Ingestion points: Project directories and files read via the
ReadandGlobtools.\n - Boundary markers: Instructions do not include boundary markers to isolate processed file content or directives to ignore instructions found within files.\n
- Capability inventory: The skill allows the use of
Bashfor command execution and theWritefor creating or modifying files.\n - Sanitization: No validation or sanitization of external data is specified.
Audit Metadata