Borrowing it
Nothing to install: this file belongs to c9r-io/orchestrator. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/c9r-io/orchestrator/main/.claude/skills/performance-testing/SKILL.mdgit clone --depth 1 https://github.com/c9r-io/orchestratorWrote 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/c9r-io/orchestrator/performance-testing)<a href="https://agentmods.dev/skills/c9r-io/orchestrator/performance-testing"><img src="https://agentmods.dev/badge/skills/c9r-io/orchestrator/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/c9r-io/orchestrator/performance-testing"><img src="https://agentmods.dev/badge/skills/c9r-io/orchestrator/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 | $0.00041 | $0.00438 |
| Opus 5.5 | $0.00016 | $0.00175 |
| Sonnet 5.5 | $0.00008 | $0.00088 |
| Haiku 4.5 | $0.00004 | $0.00044 |
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 2mo 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
Performance Testing
Benchmark the same revision, configuration, data shape, command, and transport before and after a change. Do not compare debug and release builds.
Choose a Real Surface
For this repository:
- CLI/daemon latency: time a read-only command such as
orchestrator task list -o jsonagainst the same daemon and data set. - Scheduler behavior: use deterministic mock agents under
fixtures/manifests/bundles/; do not use live AI benchmark manifests without explicit approval because they consume external credits. - Slack gateway HTTP baseline: when
orchestrator-slack-gatewayis explicitly running, its health endpoint is/healthz. - Web UI performance: use the tooling declared by
gui/package.jsonand a production build when rendering cost is the target.
Use hyperfine for command latency when installed. Otherwise run a bounded shell loop with /usr/bin/time and preserve raw samples.
HTTP Load Checks
Use hey only after the user identifies a reachable, non-production target and safe endpoint. Start with a cheap health endpoint and a low request/concurrency bound, then increase deliberately:
hey -n 200 -c 10 http://127.0.0.1:8080/healthz
Never guess API routes, credentials, or a Docker Compose environment. A generated project may carry docker/docker-compose.yml; a benchmark script is usable only when the target repository actually owns one.
Evidence
Record:
- pinned git revision and release/debug build mode
- target command or URL, payload, data size, and transport
- sample count and concurrency
- throughput, p50/p90/p99 latency, and error rate
- warm-up policy and machine/load caveats
Treat a regression as actionable only after repeating both baseline and candidate under comparable conditions. Report raw evidence and avoid universal QPS ratings detached from the tested workload.
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.
- 2mo ago First seen · 42 lines · 41 tokens per session scan A f6064dd5fe49
performance-testing is a skill published in the GitHub repository c9r-io/orchestrator (21 stars, last pushed 1mo ago), licensed MIT. It adds 41 tokens to every session and 438 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
codew-release-qa-sweep
Use before claiming Codewhale release work is done: run the full gate sweep and list the manual QA targets.
cw-dogfood
Use when a Codewhale change needs proving in the real product, or when asked to build/install/dogfood the local binaries: stamped release build, atomic install, fresh-shell verification, and the manual QA that gates cannot cover.
cw-gates
Use before claiming any Codewhale change is done, green, or ready to land: the focused-to-broad verification ladder, the budget checks CI enforces, and the rules for what counts as a passing test.
gmail
Search, read, draft, and send Gmail when the user asks to work with their inbox or email.
google-calendar
Read schedules, check availability, and create or update Google Calendar events when the user asks to work with their calendar.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.