perf-analyzer

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. It is designed to ingest and process untrusted external data such as 'Baseline data', 'Experiment results', and 'Profiling evidence'.
  • Ingestion points: Data enters the agent context through the inputs listed in the skill body.
  • Boundary markers: The instructions do not specify any delimiters (e.g., XML tags or triple quotes) to separate user data from instructions.
  • Capability inventory: No dangerous capabilities (subprocess calls, file writes, or network operations) were detected in the provided code.
  • Sanitization: There is no logic provided to sanitize or validate the external content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:48 PM