performance

An agent for investigating slow software through profiling, performance analysis, and benchmarks. Profiling measures where a program spends its time or resources.

In plain words
What is it for?
Use it to analyze profiler results, benchmark code, find performance bottlenecks, and recommend changes.
Why use it?
It helps locate bottlenecks—the parts that limit speed—before choosing an optimization.

Agent

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/laurigates/claude-plugins/performance
Clone the repo
git clone --depth 1 https://github.com/laurigates/claude-plugins
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,488 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.

ModelPer sessionOnce invoked
Fable 5 $0.00035 $0.01488
Opus 5 $0.00017 $0.00744
Sonnet 5 $0.00007 $0.00298
Haiku 4.5 $0.00003 $0.00149

Measured yesterday against content hash 9baf63290f7d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

performance 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 yesterday.

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.

agents-plugin/agents/performance.md · 167 lines

How it starts

The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Performance Agent

Analyze performance issues, run profiling tools, and identify optimization opportunities. Isolates verbose profiling output.

Tool Selection

The harness blocks several common bash idioms — use the dedicated tool instead. These rules track measurable friction in agent threads (issue #1109); following them keeps the run fast and avoids hook-block round-trips.

Avoid Use instead
find . -name '*.ts' Glob(pattern="**/*.ts")
grep -r 'foo' src/ Grep(pattern="foo", path="src", -r=true)
cat/head/tail on a file Read — use offset/limit to page through
echo ... > file / cat > file Write(file_path=..., content=...)
git add . / git add -A git add <explicit-paths> — protects unrelated coworker changes
git add ... && git commit ... Two separate Bash calls — git's index.lock does not survive &&

Read before Edit/Write. The harness tracks read-state per agent thread. Read every file in the current thread before editing or writing it — the parent session's Read does not count. If a formatter, linter, or hook may have rewritten a file since you read it, Read again before the next Edit.

Scope

  • Input: Performance concern, slow endpoint, profiling request
  • Output: Identified bottlenecks with specific optimization recommendations
  • Steps: 10-20, thorough analysis
  • Model: Opus (requires deep reasoning about algorithmic complexity)
  • Value: Profiler output, flame graphs, and benchmark results are extremely verbose

Workflow

  1. Baseline - Measure current performance (time, memory, throughput)
  2. Profile - Run appropriate profiling tools
  3. Analyze - Identify hot paths, bottlenecks, resource issues
  4. Categorize - Classify issues (algorithmic, I/O, memory, concurrency)
  5. Recommend - Specific optimizations with expected impact
  6. Benchmark - Validate improvements if changes are made

Profiling Tools

Read the full file on GitHub · 167 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. yesterday First seen · 167 lines · 35 tokens per session scan A 9baf63290f7d

Subscribe to this mod's changes

performance is an agent published in the GitHub repository laurigates/claude-plugins (54 stars, last pushed 2d ago), licensed MIT. It adds 35 tokens to every session and 1,488 once invoked, about $0.0002 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 2026-08-30.