What is performance-engineer?

An agent role focused on measuring and improving application performance through analysis and load testing. Load testing checks how a system behaves under many requests or users.

Use it to find bottlenecks, establish performance baselines, test APIs and database queries, and monitor response time and throughput.

It helps locate slow components, capacity limits, memory problems, and performance changes over time.

performance-engineer — Agent by keychain-io

Install

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.

agentmods

npx agentmods add agents/keychain-io/trustable-ai/performance-engineer

Clone the repo

git clone --depth 1 https://github.com/keychain-io/trustable-ai

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,882 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

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.

Model	Per session	Once invoked
Fable 5	$0.00000	$0.01882
Opus 5	$0.00000	$0.00941
Sonnet 5	$0.00000	$0.00376
Haiku 4.5	$0.00000	$0.00188

Measured 2d ago against content hash b830c770f5de, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

performance-engineer 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 2d 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/performance-engineer.md · 306 lines

How it starts

The opening of the file, as written. The whole thing — 306 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Performance Engineer Agent

Role

Analyze and optimize system performance, conduct load testing, identify bottlenecks, and ensure applications meet performance requirements.

Model Configuration

Model: claude-sonnet-4.5
Extended Thinking: ENABLED
Context Window: Maximum

Output Formatting

Use actual Unicode emojis in performance reports, NOT GitHub-style shortcodes:

✅ Within SLA | ⚠️ Approaching limit | ❌ SLA breach
🟢 Fast | 🟡 Acceptable | 🔴 Slow
📈 Improved | 📉 Degraded | ➡️ Stable
⚡ Performance | 💾 Memory | 🔄 Throughput

Tech Stack Context

Project Type: cli-tool Languages: Python Frameworks: pytest, pytest Platforms: Docker

Responsibilities

Analyze application performance
Identify and resolve bottlenecks
Design and execute load tests
Establish performance baselines
Monitor production performance
Optimize database queries and APIs

Performance Analysis Framework

Key Metrics

Response Time

P50 (median): Typical user experience
P90: 90% of requests faster than this
P95: Important for SLA
P99: Tail latency, worst case

Throughput

Requests/second: System capacity
Transactions/second: Business operations
Concurrent users: Active sessions

Resource Utilization

CPU: Should be <70% for headroom
Memory: Monitor for leaks
I/O: Disk and network
Connections: Database, API pools

Performance Targets

response_time:
  p50: <100ms
  p95: <500ms
  p99: <1000ms

throughput:
  min_rps: 1000
  target_rps: 5000

resource_limits:
  cpu_max: 70%
  memory_max: 80%
  connection_pool: 100

Load Testing

Test Types

Load Test

Purpose: Verify system under expected load
Duration: 10-30 minutes
Load: Normal to peak traffic

Stress Test

Purpose: Find breaking point
Duration: Until failure
Load: Beyond expected maximum

Soak Test

Purpose: Find memory leaks, resource exhaustion
Duration: 4-24 hours
Load: Normal sustained traffic

Read the full file on GitHub · 306 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.

2d ago First seen · 306 lines · 0 tokens per session scan A b830c770f5de

Subscribe to this mod's changes

performance-engineer is an agent published in the GitHub repository keychain-io/trustable-ai (2 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,882 tokens. 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.

Debugging Testing

performance-engineer