file-operations
Pass
Audited by Gen Agent Trust Hub on Nov 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes instructions for destructive operations such as recursive directory deletion and batch file removal. Evidence: Commands like
rm -rf dirandfind /path -name "*.tmp" -deletein SKILL.md. - [PRIVILEGE_ESCALATION]: The Troubleshooting section suggests using elevated privileges to resolve permission issues. Evidence: Mention of using
sudofor "Permission denied" errors in SKILL.md. - [INDIRECT_PROMPT_INJECTION]: The skill provides instructions that ingest untrusted filesystem data, creating a potential injection surface. 1. Ingestion points: Filenames and file contents are accessed via
find,ls,cat,head, andtail(SKILL.md). 2. Boundary markers: No delimiters or "ignore embedded instructions" warnings are specified for the processed data. 3. Capability inventory: Modification and deletion capabilities includingrm,chmod,chown, andsed -iare present in SKILL.md. 4. Sanitization: No validation or sanitization of filenames or content is described before use in subsequent commands.
Audit Metadata