awesome-performance-audit
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting and analyzing potentially untrusted external data, such as source code, configuration files, system profiles, and traces, which creates a surface for indirect prompt injection attacks.
- Ingestion points: Source code files, system configuration (timeouts, limits), heap snapshots, CPU profiles, and garbage collection traces as specified in the "Scope and method" and "Track" sections of SKILL.md.
- Boundary markers: The instructions do not mandate the use of specific delimiters or protective markers (e.g., "ignore embedded instructions") when processing audited content.
- Capability inventory: The skill leverages file read capabilities and system diagnostic commands (uptime, dmesg, vmstat, mpstat, pidstat, iostat, free, sar, top) to collect evidence.
- Sanitization: There is no evidence of sanitization or content validation for the data ingested during the audit process.
- [COMMAND_EXECUTION]: The skill provides explicit instructions for the agent to execute a variety of system diagnostic and performance monitoring commands.
- Evidence: The "Triage before reading code" section in SKILL.md lists uptime, dmesg -T | tail, vmstat 1, mpstat -P ALL 1, pidstat 1, iostat -xz 1, free -m, sar -n DEV 1, sar -n TCP,ETCP 1, and top as primary tools for resource measurement.
Audit Metadata