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 shandar/pwp-plugin --skill pwp-perfgit clone --depth 1 https://github.com/shandar/pwp-pluginWrote 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/shandar/pwp-plugin/pwp-perf)<a href="https://agentmods.dev/skills/shandar/pwp-plugin/pwp-perf"><img src="https://agentmods.dev/badge/skills/shandar/pwp-plugin/pwp-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/shandar/pwp-plugin/pwp-perf"><img src="https://agentmods.dev/badge/skills/shandar/pwp-plugin/pwp-perf.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 | $0.00113 | $0.00900 |
| Opus 5.5 | $0.00045 | $0.00360 |
| Sonnet 5.5 | $0.00023 | $0.00180 |
| Haiku 4.5 | $0.00011 | $0.00090 |
Grade A, and why
pwp-perf 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 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.
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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Performance Optimization Skill
The cardinal rule: never optimize on vibes. Measure, identify the bottleneck, fix it, prove it's faster.
Performance Mindset
- Measure before optimizing. Intuition about performance is wrong more often than right. Profile first.
- Optimize the bottleneck. Making a fast operation faster doesn't help.
- Premature optimization is real. Only optimize measured problems or known-critical paths.
- Verify the improvement. Before/after numbers or it didn't happen.
Investigation Protocol
Step 1: Define the Problem
- What is slow? (Page load, API, render, build)
- How slow? (Actual numbers: "3.2 seconds")
- What is the target? ("Under 1 second")
Step 2: Measure Current State
| Area | How to Measure |
|---|---|
| Page load | Lighthouse, DevTools Performance tab |
| Core Web Vitals | LCP, INP, CLS via Lighthouse |
| API response | Network tab timing, server-side logging |
| Render | React DevTools Profiler |
| Bundle size | Build output, bundlephobia |
| Memory | DevTools Memory tab, heap snapshots |
| Database | Query EXPLAIN plans |
Step 3: Identify the Bottleneck
- Network — too many requests, large payloads, no caching
- Data — slow queries, over-fetching, N+1 patterns
- Rendering — unnecessary re-renders, layout thrashing
- Compute — expensive calculations on main thread
- Assets — uncompressed images, unminified JS
Step 4: Fix the Bottleneck
| Bottleneck | Common Fixes |
|---|---|
| Large bundle | Code splitting, dynamic imports, tree shaking |
| Slow page load | Lazy loading, preload critical resources |
| Re-renders | useMemo, useCallback, React.memo (only where profiled) |
| Slow API | Caching, pagination, field selection |
| N+1 queries | Batch queries, eager loading, DataLoader |
| Large images | WebP/AVIF, responsive srcset, lazy loading |
| Heavy computation | Web Workers, caching results |
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 First seen · 85 lines · 113 tokens per session scan A 03735325080a
pwp-perf is a skill published in the GitHub repository shandar/pwp-plugin (1 stars, last pushed 7mo ago), licensed MIT. It adds 113 tokens to every session and 900 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
tang-rou
Ative TANG-ROU (Soft Mist) quando o usuário nomear TANG-ROU ou pedir desenvolvimento, correção, auditoria, otimização ou planejamento de Foundry VTT e programação aplicada: macros, módulos, sistemas, CSB, Combat Tracker Dock, Midi-QOL, DAE, Sequencer, Warp Gate, Active Effects, hooks, compêndios, fichas homebrew, erro…
watch-errors
Quiet operator watch for a detached clawback proxy — stream ONLY legitimate errors and never the benign flood. Wakes on [error]/[fatal] lines, crash signatures (uncaught/EADDRINUSE/ECONN/Error:/self-loop), server-side 5xx that are NOT the transient 529, real (non-keep-alive) request failures, and a loud "PROXY DOWN"…
inspect
Statically inspect prompt-cache breakpoints in a captured /v1/messages body or a clawback --state file, with ZERO token spend and no API calls. Answers "where are the cachecontrol breakpoints, and which ephemeral tokens (cch / date / ) sit inside which cached prefixes?" — i.e. what a single mutation cold-starts. Use…
monitor
Run the clawback proxy and watch its log output for errors. Starts the bare proxy (honoring config layering), captures combined stdout+stderr to a log file, and follows it with warn/error/4xx/5xx lines highlighted. Supports foreground follow, background --detach (with --stop/--status), and --attach to tail an…
operate
Watch an already-detached clawback proxy as an event stream — surface only actionable log lines (error/warn/4xx/5xx/crash signatures, plus rebind and runtime-mode toggles) AND announce a loud "PROXY DOWN" line the instant the proxy pid dies. Fills the gap that runmonitor.sh --attach leaves: a bare tail goes SILENT on…
verify-config
Config doctor for clawback — loads the merged config (DEFAULTS < global < ./CLAWBACK.md < CLI overrides) exactly as the proxy's loadConfig does, prints a secret-free summary of the resolved values plus the merge-order sources, and exits non-zero if the canonical CLAWBACK.md did NOT take effect (host not 0.0.0.0, tls…