extreme-software-optimization

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references several well-known third-party profiling and optimization tools. It provides instructions to install utilities such as clinic, 0x, py-spy, snakeviz, and scalene using standard package managers (npm, pip). These are widely recognized developer tools used for legitimate performance analysis.
  • [COMMAND_EXECUTION]: The instructions guide the agent in using standard system diagnostics and profiling tools, including perf, strace, valgrind, heaptrack, and hyperfine. These commands are executed locally to gather performance metrics and identify bottlenecks, which is the primary purpose of the skill.
  • [DATA_EXFILTRATION]: A curl command is mentioned for retrieving trace data from a local Go pprof server (localhost:6060). This is a standard procedure in Go performance profiling and does not involve sending data to external or untrusted domains.
  • [REMOTE_CODE_EXECUTION]: No patterns of remote code execution or fetching and executing untrusted scripts were found. All execution patterns are directed at local binaries, standard compilers, or reputable package manager registries.
  • [CREDENTIALS_UNSAFE]: The skill does not contain hardcoded secrets, nor does it attempt to access sensitive files like SSH keys, AWS credentials, or environment configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 11:08 AM
Security Audit — agent-trust-hub — extreme-software-optimization