The Agent Skills Directory

[SAFE]: The skill provides procedural guidance for performance engineering. No malicious patterns, obfuscation, or unauthorized data access were detected.
[SAFE]: The instructions contain explicit security guardrails, specifically warning the agent never to include prompts, identities, IDs, or credentials in fixtures, logs, or artifacts during benchmark creation.
[SAFE]: References to external project documentation and related internal skills (e.g., ci-test-reliability, pre-push-checks) are within the expected scope of a development environment and do not indicate malicious data exfiltration or unauthorized access.

dsh-speed-up-perf