fsiter
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
x fsitercommand to perform file system operations such as listing files (--ls), counting items (--filecount), and checking if directories are empty (--dirempty). These are standard administrative tasks. - [COMMAND_EXECUTION]: The
--dfs(depth-first search) flag allows specifying a callback command (e.g.,echo) to be executed for each discovered item during traversal. While this allows for repetitive command execution, the examples provided demonstrate benign use cases for automation and reporting. - [DATA_EXFILTRATION]: The skill is designed to iterate through file systems and retrieve metadata (counts, existence, names). While it accesses directory structures, there are no network components or sensitive file access patterns that suggest unauthorized data transmission.
- [INDIRECT_PROMPT_INJECTION]: The skill functions as a data ingestion surface by reading file and directory names from the environment. If an agent processes these names as instructions, it could be vulnerable to indirect prompt injection from malicious filenames. This is an inherent risk of file system tools and requires the agent to treat tool outputs as untrusted data.
Audit Metadata