performance

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and profile external C++ source files and binaries, which presents a potential indirect prompt injection surface.
  • Ingestion points: The scripts/perf_analyzer.py script reads source code from files provided as CLI arguments, and the profiling workflow involves running tools on local binaries.
  • Boundary markers: There are no explicit instructions or delimiters used to ensure the agent ignores embedded instructions within the user-provided code comments or strings.
  • Capability inventory: The skill workflow includes the execution of local binaries via profiling tools such as perf and valgrind. While the analyzer script itself is limited to regex matching, the overall process requires running the code being optimized.
  • Sanitization: The skill does not implement sanitization or verification procedures for the C++ code or binaries before they are processed or executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 02:37 AM