profiling-qwen3-tts
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides commands to run performance benchmarks within a Docker container using
docker runandcargo run. It also executes a local shell scriptscripts/audit-gpu-syncs.shto identify GPU synchronization bottlenecks. These operations are consistent with the skill's stated purpose of performance profiling and optimization. - [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection because it processes text data for TTS synthesis during benchmarks. However, the risk is negligible as the data is used for measuring execution time and resource usage of a non-LLM engine.
- Ingestion points: Text inputs for synthesis benchmarks defined in SKILL.md.
- Boundary markers: No specific delimiters or safety warnings for embedded instructions are used in the benchmarking command examples.
- Capability inventory: The skill utilizes host-level execution of
docker,cargo, andbashfor benchmarking and auditing. - Sanitization: No explicit sanitization of the input text is mentioned, as the primary goal is performance measurement.
Audit Metadata