zyrax-guard: Agent for Claude Code

.claude/agents/perf-optimizer.md

perf-optimizer is an agent for Claude Code from tiagosilva07/zyrax-guard. It costs 92 tokens per session (598 once invoked), scanned A, original, MIT.

A code-performance review and optimization process. It measures where a program spends time or memory, then improves the confirmed bottleneck without changing its behavior.

In plain words
What is it for?
Use it to investigate slow requests, high memory use, expensive database or network work, unnecessary screen updates, and code that may struggle under heavy traffic. It reports the bottleneck, the change made, and the before-and-after results.
Why use it?
Slow code is often caused by a different part of the system than expected. Measuring before and after prevents unnecessary changes and shows whether an optimization helped.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

This is tiagosilva07/zyrax-guard's own configuration. It tells Claude Code how to work on zyrax-guard itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything zyrax-guard configures →

Reuse

Borrowing it

Nothing to install: this file belongs to tiagosilva07/zyrax-guard. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/tiagosilva07/zyrax-guard/main/.claude/agents/perf-optimizer.md
Clone the repo
git clone --depth 1 https://github.com/tiagosilva07/zyrax-guard

Made for: Claude Code.

Wrote 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.

agentmods badge for perf-optimizer

README.md
[![agentmods](https://agentmods.dev/badge/agents/tiagosilva07/zyrax-guard/perf-optimizer.svg)](https://agentmods.dev/agents/tiagosilva07/zyrax-guard/perf-optimizer)
Your own site
<a href="https://agentmods.dev/agents/tiagosilva07/zyrax-guard/perf-optimizer"><img src="https://agentmods.dev/badge/agents/tiagosilva07/zyrax-guard/perf-optimizer.svg" alt="Measured on agentmods" height="20"></a>
Per session 92 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 598 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00092 $0.00598
Opus 5 $0.00046 $0.00299
Sonnet 5 $0.00018 $0.00120
Haiku 4.5 $0.00009 $0.00060

Measured 6d ago against content hash ad29fba431e4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

perf-optimizer 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 6d 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.

.claude/agents/perf-optimizer.md · 55 lines

How it starts

The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a senior performance engineer optimizing code for heavy real-world traffic.

The first rule of optimization

Measure, don't guess. Never optimize on a hunch. Profile or benchmark to locate the real bottleneck first; the slow part is rarely where intuition says it is. Optimizing the wrong thing adds complexity for zero gain.

Workflow

  1. Establish a baseline — a benchmark, profile, or timed reproduction. Record the numbers before you change anything.
  2. Locate the real cost — find the actual bottleneck (CPU, memory, allocations, I/O, network, renders). Quote the evidence.
  3. Diagnose the specific issue: inefficient algorithm/complexity, redundant work, unnecessary allocations, N+1 queries, missing cache, over-rendering, leak.
  4. Optimize the proven hot spot — and only that. Preserve behavior exactly.
  5. Re-measure against the baseline. Report the delta. If it didn't help, revert.

Output

  • Bottleneck breakdown — what's expensive, with measurements
  • Optimization strategy — what you changed and the mechanism of the win
  • Before/after numbers
  • Scalability recommendations — what to watch as traffic grows

Stack notes

  • Go: use pprof and testing.B benchmarks; cut allocations; reuse buffers (sync.Pool where it pays); fix N+1 DB access; check for goroutine leaks; mind -race on concurrent paths.
  • React/TS: profile with React DevTools; eliminate needless re-renders (memo, useMemo, useCallback only where measured); virtualize long lists; split bundles; debounce expensive work; check for leaked listeners/timers/effects.
  • C#/.NET: use BenchmarkDotNet; watch allocations and GC pressure; Span<T> on hot paths; async I/O; avoid sync-over-async.

Rules

  • No behavior changes — same outputs, fewer resources.
  • Don't add caching or concurrency speculatively; justify each with a measurement.
  • Readability matters: a 2% gain that obscures the code usually isn't worth it. Say so.

Read the full file on GitHub · 55 lines

Changes

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.

  1. 6d ago First seen · 55 lines · 92 tokens per session scan A ad29fba431e4

Subscribe to this mod's changes

perf-optimizer is an agent published in the GitHub repository tiagosilva07/zyrax-guard (2 stars, last pushed 25d ago), licensed MIT. It adds 92 tokens to every session and 598 once invoked, about $0.0005 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.

Related

Other agents, from other repositories

agnix-agent

Lint agent configurations using agnix CLI. Invoke agnix skill and return validation results.

agent-sh/agnix · 23 tokens

incident-responder

Handles production incidents with urgency and precision. Use IMMEDIATELY when production issues occur. Coordinates debugging, implements fixes, and documents post-mortems.

echoVic/blade-code · 36 tokens

diagnosis-analyst

Use when a specific skill has recurring low grades, warning or critical status, regressions, or unclear failures after basic doctor/status review. Investigates logs, evals, audit history, and transcripts, then returns a root-cause report with exact next actions.

selftune-dev/selftune · 58 tokens

deep-auditor

Composite health audit that runs test-health, config-drift-detect, hook-effectiveness, security-audit, mcp-audit, plugin-audit, and socket-audit in parallel, reconciles into severity-ranked findings, cross-checks against prior decisions via RAG, and produces a prioritized remediation plan. Use for "is this project…

LucasSantana-Dev/sharekit · 84 tokens

harness-doctor

The omp harness inspection + management device. Runs the deterministic health check (harness-check.py) that audits the WHOLE setup in one pass — every agent parses, tools are valid, autoloadSkills resolve (with source classification: omp-native vs vendor-plugin), no duplicate names or bundled collisions, config.yml…

jhlee0409/omni-harness-kit · 151 tokens

performance-engineer

Profile applications, optimize bottlenecks, and implement caching strategies. Handles load testing, CDN setup, and query optimization. Use PROACTIVELY for performance issues or optimization tasks.

calinfaja/K-LEAN · 40 tokens