python-performance-optimization
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious intent or security concerns were detected. The skill follows best practices for the domain of software performance tuning.
- [EXTERNAL_DOWNLOADS]: Recommends the use of widely accepted community packages for performance analysis. Evidence: mentions of installing line-profiler, memory-profiler, py-spy, and pytest-benchmark.
- [COMMAND_EXECUTION]: Provides instructions for using standard profiling CLI tools to analyze script behavior. Evidence: usage of python -m cProfile, kernprof, and py-spy to profile applications.
- [DATA_EXFILTRATION]: Network code examples target a well-known service intended for development and testing. Evidence: code snippets performing GET requests to https://httpbin.org/delay/1.
- [PROMPT_INJECTION]: The skill's primary function of profiling user code creates an inherent indirect prompt injection surface.
- Ingestion points: Reading local scripts and data files via open() and file iteration.
- Boundary markers: None present.
- Capability inventory: High capability to execute code for profiling via cProfile, timeit, and external process management.
- Sanitization: None present.
Audit Metadata