Performance Deity — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Performance Deity (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
Aggregate score unchanged between these scans.
The primary manifest — the file an agent reads to learn what this artifact does.
A performance engineering plugin for Claude Code — and a `CLAUDE.md` template for any AI coding assistant.
Nine benchmark-driven skills that enforce a strict, phased performance methodology. Every skill requires proof before presenting a change: a before/after table with real numbers. No guessing.
/plugin marketplace add https://github.com/v0idOS/performance-deity
/plugin install performance-deity@performance-deityThis gives you nine slash commands in Claude Code:
| Command | What it does |
|---|---|
/optimize | Benchmarks, analyzes Big-O, rewrites, proves improvement |
/memory | Instruments memory, proves the leak grows, fixes the reference |
/concurrency | Fires 1k–10k concurrent requests, proves failure, adds atomic ops |
/database | Runs EXPLAIN ANALYZE, kills N+1 loops, provides CREATE INDEX statements |
/stress-test | Bombards code with extreme inputs, documents crashes, hardens the boundary |
/network | Profiles payload bytes, enforces Brotli/ETag, migrates to GraphQL/protobuf |
/build | Fixes Docker layer order, Webpack caching, GitHub Actions pipelines |
/telemetry | Wraps code in OpenTelemetry spans, proposes concrete alert rules |
/ui | Audits React re-renders, GPU-accelerates animations, virtualizes long lists |
If you're not using Claude Code's plugin system, copy CLAUDE.md directly into your project root. Claude Code reads it automatically on every session — no flags, no config.
git clone https://github.com/v0idOS/performance-deity
cp performance-deity/CLAUDE.md /your/project/CLAUDE.mdFor Cursor or Windsurf:
cp performance-deity/CLAUDE.md /your/project/.cursorrulesUnlike other plugins that require you to copy a tools/ folder into your project, Performance Deity generates its own benchmarking tools on the fly.
Whenever Claude needs to prove a latency reduction, it writes an ephemeral micro-benchmark script in the target language (Node.js, Python, Rust, etc.), runs it to collect P95 and Average execution times (including a warm-up phase to prime JIT caches), and then deletes the script. No clutter, no dependencies.
Blocks commits containing unresolved TODO: optimize markers:
cp hooks/pre-commit .git/hooks/pre-commit
chmod +x .git/hooks/pre-commit.
├── CLAUDE.md ← Drop-in template for any AI assistant
├── README.md
├── LICENSE ← MIT
├── .claude-plugin/
│ ├── plugin.json ← Plugin manifest (Claude Code plugin system)
│ └── marketplace.json ← Marketplace catalog (buildwithclaude.com)
├── skills/ ← Slash command skill files (Claude Code plugin)
│ ├── optimize/SKILL.md
│ ├── memory/SKILL.md
│ ├── concurrency/SKILL.md
│ ├── database/SKILL.md
│ ├── stress-test/SKILL.md
│ ├── network/SKILL.md
│ ├── build/SKILL.md
│ ├── telemetry/SKILL.md
│ └── ui/SKILL.md
├── hooks/
│ └── pre-commit ← Optional git hook
└── archive/ ← Original SKILL.md files (pre-refactor)CLAUDE.mdmarketplace.json and slash commands.cursorrules variant maintained in sync with CLAUDE.mdBuilt for developers who demand blazing-fast software.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.