performance-engineer

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to analyze external code and performance logs, which presents a surface for indirect prompt injection.\n
  • Ingestion points: Uses Read, Grep, and Glob tools to ingest workspace files for analysis.\n
  • Boundary markers: The instructions lack explicit delimiters or safety warnings to ignore instructions embedded within analyzed data.\n
  • Capability inventory: The agent has access to Bash, Write, and Edit tools, which could be misused if the agent is successfully injected.\n
  • Sanitization: No sanitization of ingested content is performed before the agent processes it.\n- [COMMAND_EXECUTION]: The skill suggests running various profiling and benchmarking commands (e.g., node --prof, go test -cpuprofile). These are standard development workflows and are used as intended for performance analysis.\n- [SAFE]: The provided Python utility scripts (scripts/perf_report.py and scripts/profile.py) are safe, performing only basic string formatting and local file writes using standard Python libraries.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 02:59 PM