remove-pauses-from-track
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill coordinates multiple shell commands using
premiere-cliandremove-pausesto perform project introspection, audio extraction, and sequence manipulation. These tools are used to process files saved in the/tmp/directory. - [DYNAMIC_EXECUTION]: The instructions include a Python code template in Step 4 that the agent must execute in the
premiere-proenvironment. This snippet uses thepremiere_ailibrary to verify the results of the audio processing by detecting remaining silence. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Premiere Pro sequences, creating a potential surface for indirect prompt injection.
- Ingestion points: Audio data exported from Premiere Pro sequences (
.wavfiles) and sequence metadata (name and frame rate). - Boundary markers: Absent. The skill does not define delimiters or specific warnings to ignore embedded instructions within the audio content.
- Capability inventory: The skill has the ability to modify Premiere Pro timelines via
remove-track-intervals, execute shell commands, and run Python code. - Sanitization: Absent. There is no explicit validation or escaping of the audio content or sequence names before they are processed by the CLI tools or interpolated into the verification script.
Audit Metadata