Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/new1direction/korgex/performance-profilingnpx skills add New1Direction/korgex --skill performance-profilinggit clone --depth 1 https://github.com/New1Direction/korgexWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00017 | $0.00314 |
| Opus 5 | $0.00009 | $0.00157 |
| Sonnet 5 | $0.00003 | $0.00063 |
| Haiku 4.5 | $0.00002 | $0.00031 |
Grade A, and why
performance-profiling scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
The cardinal rule: measure before you optimize, and measure again after. Intuition about what's slow is usually wrong.
- Define "fast enough". What's the target (latency, throughput, memory) and on what input? Without a target you can't know when to stop.
- Reproduce + measure the baseline. Get a repeatable workload and time it. Record the number — it's what you'll compare against.
- Profile to find the real hotspot. Use a profiler / timing instrumentation, not eyeballing. Find where the time/allocations actually go. The bottleneck is often not where you'd guess (it's frequently I/O, N+1 queries, or an accidental O(n²), not "slow code").
- Fix the biggest cost first. Often algorithmic (data structure, caching, batching) beats micro-optimization. Change one thing.
- Re-measure. Confirm the change actually helped against the baseline. Keep it only if it did; revert if it didn't.
- Guard it. For a critical path, add a benchmark/regression check so it can't silently regress later.
Red flags: optimizing without a measurement, micro-tuning a cold path, trading readability for speed that doesn't move the target number.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 28 lines · 17 tokens per session scan A 5f0d9170d732
performance-profiling is a skill published in the GitHub repository New1Direction/korgex (5 stars, last pushed 2mo ago), licensed MIT. It adds 17 tokens to every session and 314 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
vibe-ui-design
Use when designing UI/UX, defining design systems, choosing color palettes, typography, spacing, or component primitives. Triggers on "design system", "design tokens", "color palette", "UI 设计", "深色工业风", "design language".
vibe-security
Use when performing security audits, vulnerability scanning, dependency checks, or secure code reviews. Triggers on "security audit", "scan vulnerabilities", "安全审查", "漏洞扫描", "OWASP", "dependency check".
vibe-ui-beautify
Use when adding animations, micro-interactions, 3D effects, glassmorphism, or "juicy" tactile feedback to UI components. Triggers on "animation", "hover effect", "transition", "动效", "果汁感", "Framer Motion", "GSAP", "3D 视效".
vibe-plan
Use this skill to create a phased development plan based on requirements, architecture, and API docs.
vibe-test
Use when writing tests, performing unit testing, integration testing, or verifying system functionality. Triggers on "write tests", "run tests", "add unit test", "TDD", "测试", "跑测试".
vibe-autopilot
Use this skill when the user requests fully autonomous, 24/7 hands-free development without human intervention.