smart-file-organizer
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes the contents of various untrusted file types (TXT, PDF, DOCX, HWP, HWPX) to generate descriptive filenames. This presents an indirect prompt injection surface where a malicious file could contain instructions designed to influence the agent's behavior during the organization process. Ingestion points: SKILL.md Step 3 (Analysis of text-based files, documents, and images). Boundary markers: None identified. The skill does not instruct the agent to use delimiters or ignore instructions found within processed files. Capability inventory: File renaming (mv), directory creation (mkdir), directory removal (rmdir), and executing local scripts (scripts/rename-map.sh) that use python3. Sanitization: No evidence of content sanitization or instruction filtering before processing file data.
- [COMMAND_EXECUTION]: The skill utilizes a local shell script (scripts/rename-map.sh) to manage file rename maps and perform rollbacks. This script involves operations like mv, mkdir, and python3 -c for JSON processing. While these operations are consistent with the skill's stated purpose, they represent a significant capability that could be exploited if the agent's logic is compromised by malicious input.
Audit Metadata