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 OutThisLife/brooklyn-skills --skill perfgit clone --depth 1 https://github.com/OutThisLife/brooklyn-skillsWrote 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/outthislife/brooklyn-skills/perf)<a href="https://agentmods.dev/skills/outthislife/brooklyn-skills/perf"><img src="https://agentmods.dev/badge/skills/outthislife/brooklyn-skills/perf/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/outthislife/brooklyn-skills/perf"><img src="https://agentmods.dev/badge/skills/outthislife/brooklyn-skills/perf.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00042 | $0.02722 |
| Opus 5.5 | $0.00017 | $0.01089 |
| Sonnet 5.5 | $0.00008 | $0.00544 |
| Haiku 4.5 | $0.00004 | $0.00272 |
Grade A, and why
perf scanned grade A with 1 finding 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Verify the delivered page's boot entry before attributing a disappearing CMS slot to async registration. In Vue, `createSSRApp` preserves existing SSR nodes while an async component loads; `createApp` clears the mount How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Perf
The general performance loop. Profiling (CPU profile, flamegraph, memory snapshot, trace) is the usual tool, but the loop is the same regardless of which one fits.
Loop
- Reproduce the slow path and capture a baseline number — time, memory, FPS, request latency, whatever matters here.
- Profile to find the real hot path: CPU profile / flamegraph / memory snapshot / trace as fits. Reuse the repo's existing profiler; don't build a parallel harness.
- Read the profile, not your intuition — fix the actual hot path.
- Re-measure. Record before/after in the PR title/description.
- Ship in topical commits (
pr-update),clean, keep CI green.
Live Electron / hgui
- Attach to the existing renderer with the repository's CDP helper and exact dev-server URL; preserve tabs, drafts and running agents. Keep source edits in the worktree and save reversible live injections outside the app's Tailwind source scan.
- Separate uninstrumented frame timing from CPU, selector-statistics and React attribution passes. Do not run your own tests/builds concurrently with comparison captures. Record mounted/visible panes and busy-session counts with each sample; persist raw samples before reporting tables.
- Treat geometry/read functions atop a CPU profile as possible flush victims. Trace
UpdateLayoutTreeand style invalidations before optimizing their callers. Isolate interacting selector families together: removing one broad:has()rule can show no gain while other rules still invalidate the same subtree. Preserve behavior with positively anchored owner selectors, sibling rules, or existing observed state; do not ban:has()generally. - Recheck target ID, code revision, runtime state and injection globals after interruption/restart. A fresh process or independently updated checkout is a new baseline, never an after result for the old renderer. Report aged and fresh results separately; native/embedder heap growth or a GC pause alone does not prove a leak.
- Put lazy render callbacks below existing presence boundaries.
items(kit)eagerly creates a closed menu's elements even when its portal mounts nothing; a small child component lets presence defer construction without replacing open/close mechanics. Test both dropdown and context paths plus latest-item selection. - Verify Vite Fast Refresh's actual component family IDs before injection: local registrations and
exportregistrations are distinct, and registering an already-known type under a second ID is ignored. Use the shippedvalidateRefreshBoundaryAndEnqueueUpdateexport whereperformReactRefreshis private, and verify rendered markers afterward. - Detect live render failures from boundary state or error elements, not whole-body text: this conversation's code/tool output can itself contain the error phrase. Restore instrumentation in
finally, and describe live injections as reload-volatile.
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.
- 1mo ago Changed · +55 lines 5b2abfe13379
- 2mo ago First seen · 32 lines · 42 tokens per session scan A 7f89136c7030
perf is a skill published in the GitHub repository OutThisLife/brooklyn-skills (209 stars, last pushed 7d ago), licensed MIT. It adds 42 tokens to every session and 2,722 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
cwv-optimizer
Diagnose and fix Core Web Vitals issues on AEM Edge Delivery Services pages. Goes deeper than generic CWV advice by understanding EDS-specific performance patterns including the 100KB LCP budget, E-L-D loading phases, block rendering behavior, and third-party script impact. Produces specific fixes for LCP, CLS, and…
incident-response-incident-response
Coordinate production incident triage, mitigation, investigation, recovery, and follow-up. Use when responding to outages or other incidents that require structured operational coordination.
ci-fixer
CI failures - read error, minimal fix, verify.