performance-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute performance comparison commands using
pnpm(e.g.,pnpm runtimeperf-compare). These are standard development tasks intended for local benchmarking within the project environment. - [INDIRECT_PROMPT_INJECTION]: The skill involves reading local project files such as
README.mdand benchmark scripts underpackages/orbenchmark/to determine execution parameters. While this presents a minor attack surface if an attacker can modify project files, it is a standard workflow for performance analysis and is considered a low risk in this context. - [SAFE]: No remote code execution, data exfiltration, obfuscation, or malicious prompt injection patterns were detected. The skill's operations are confined to the local repository and standard development tools.
Audit Metadata