silence-cut-edl
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the
auto-editorcommand-line utility. This is the primary and intended purpose of the skill, facilitating audio processing through a well-known third-party tool. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes user-provided file paths and parameters.
- Ingestion points: The skill takes several inputs from the user, including
Input file,Editor target,Threshold,Margin, andOutput pathinSKILL.md. - Boundary markers: The instructions use double quotes around variables in the shell command (e.g.,
"<input>") which helps prevent issues with spaces but does not fully sanitize against malicious shell character injection. - Capability inventory: The skill utilizes the agent's ability to run shell commands (via
bash) to invokeauto-editor. - Sanitization: No explicit validation or sanitization of the input strings is mandated by the instructions, relying on the agent's underlying safety filters and the shell's handling of quoted strings.
Audit Metadata