Audit Performance
SkillDev toolsUse when reviewing or auditing tenferro-rs code for performance rule violations: before merging changes that touch tensor kernels, graph/compiler planning, caches, GPU kernels, Faer integration, benchmarks, or examples, when a workload is unexpectedly slow, or when scanning the repository for latent performance anti-patterns. Static audit that reports violations of PERFORMANCE_TIPS.md with file and line and never claims a speedup or slowdown without measurement.
Available today. Use it from your connected AI after setup.
No other account needed.
Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
Then ask your AI: use the Audit Performance skill
What this skill tells your AI
The instructions your AI receives, as published by tensor4all/tenferro-rs in .agents/skills/audit-performance/SKILL.md and read by ahel’s review.
This is a thin launcher; it carries no rule content.
- Read
PERFORMANCE_TIPS.mdin full. ItsAudit Proceduresection defines the scope handling, theDetect/Fixhints, and the report format. - Take the argument as the scope:
fullfor the whole repository, otherwise the given paths. Without an argument, audit the current diff againstorigin/main. - Follow the audit procedure and report each finding as
file:line, thePERFORMANCE_TIPS.mdsection title, the evidence, and the remediation direction. - Findings are static rule violations, not measured regressions. Do not claim
a speedup or slowdown, and route any proposed optimization through the
Performance-Gated Experiment Protocol in
PERFORMANCE_TIPS.md.
Signals
- GitHub stars
- 79
- Forks
- 4
- Last commit
- Sep 2026
Advanced
- Item type
- skill
- Key
audit-performance-tensor4all- Source
- github.com/tensor4all/tenferro-rs