The Agent Skills Directory

[SAFE]: The skill focuses on performance testing methodologies and provides documentation for standard industry tools. No malicious patterns such as prompt injection or behavioral overrides were detected in the instructions.
[SAFE]: The core utility script scripts/perf_budget_checker.py is implemented using only the Python standard library. It performs local file reading and basic mathematical comparisons without network access, subprocess spawning, or unsafe dynamic code execution.
[SAFE]: External resources and documentation links in data/sources.json target trusted organizations and well-known services, including Google (web.dev), Grafana (k6), and official language diagnostics (Node.js). These are used for informational purposes and do not introduce unverifiable dependencies.
[SAFE]: Data ingestion in the budget checker script is limited to parsing local JSON files via the standard json module. There is no evidence of the data being used to influence agent prompts or system commands.
[SAFE]: No persistence mechanisms, privilege escalation attempts, or obfuscated content were found across the skill files.

qa-testing-performance