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 skills add pnakhat/qa-ai-repo --skill performance-testinggit clone --depth 1 https://github.com/pnakhat/qa-ai-repoWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/pnakhat/qa-ai-repo/performance-testing)<a href="https://agentmods.dev/skills/pnakhat/qa-ai-repo/performance-testing"><img src="https://agentmods.dev/badge/skills/pnakhat/qa-ai-repo/performance-testing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/pnakhat/qa-ai-repo/performance-testing"><img src="https://agentmods.dev/badge/skills/pnakhat/qa-ai-repo/performance-testing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00159 | $0.03360 |
| Opus 5 | $0.00079 | $0.01680 |
| Sonnet 5 | $0.00032 | $0.00672 |
| Haiku 4.5 | $0.00016 | $0.00336 |
Grade A, and why
performance-testing 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 8d 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.
How it starts
The opening of the file, as written. The whole thing — 218 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Performance Testing
Performance testing answers two questions with numbers, not vibes: "is it fast enough?" and "how much load before it breaks?" You cannot answer either without a target. So the work is always in this order: define the SLOs, model a realistic workload, express the thresholds as gates, run against a production-like target, then read the percentiles. A run with no SLO is a benchmark nobody can fail — it decorates a dashboard and changes no decision.
SLOs first — the metrics that matter
Decide the target before you write a script. Every threshold in the test traces back to one of these. Report the whole distribution; the average is the metric that lies.
| Metric | What it is | Target guidance |
|---|---|---|
| p95 / p99 latency | 95th/99th-percentile response time — 1 in 20 / 1 in 100 requests is slower | Set per endpoint from user expectation, e.g. read API p95 < 300 ms, p99 < 800 ms |
| Error rate | share of requests that fail (5xx, timeouts, failed checks) at load | < 0.1% under expected peak; a fast error is still an error |
| Throughput (RPS) | requests/sec the system sustains within SLO | Derive from real traffic; it's a result, not a knob |
| Saturation | how full the resource is (CPU, memory, connection pool, queue depth) | Alert well before 100% — the knee is near saturation |
| LCP (frontend) | Largest Contentful Paint — main content visible | "Good" ≤ 2.5 s at p75 of real users |
| INP (frontend) | Interaction to Next Paint — responsiveness | "Good" ≤ 200 ms at p75 |
| CLS (frontend) | Cumulative Layout Shift — visual stability | "Good" ≤ 0.1 at p75 |
Why averages lie. Mean latency is dominated by the fast majority and erased by a few extreme values. A service with a 40 ms mean can still have a 2 s p99 — and that p99 is the checkout that times out, the customer who leaves. If a request has 5 downstream calls, the slowest of the five sets the response, so tail latency compounds across a request path. Always report p50/p95/p99 (and the max); quote a mean only alongside them, never instead.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 218 lines · 159 tokens per session scan A 83a7bb0709c5
performance-testing is a skill published in the GitHub repository pnakhat/qa-ai-repo (2 stars, last pushed 2mo ago), licensed MIT. It adds 159 tokens to every session and 3,360 once invoked, about $0.0008 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
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.