performance

An AI software role focused on finding and improving slow parts of a program by measuring them before and after changes. It examines speed, memory use, input/output, allocations, and concurrent work.

In plain words
What is it for?
Use it to benchmark, profile, or optimize a chosen target for latency, throughput, or memory, then report the bottleneck and the measured before-and-after result.
Why use it?
It replaces guesses about performance problems with measured evidence and helps avoid changes that make code harder to understand without meaningful benefit.

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/wrongstack/wrongstack/performance
Clone the repo
git clone --depth 1 https://github.com/WrongStack/WrongStack
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 219 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 $0.00000 $0.00219
Opus 5 $0.00000 $0.00110
Sonnet 5 $0.00000 $0.00044
Haiku 4.5 $0.00000 $0.00022

Measured 2d ago against content hash c9cdae4d1cef, 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 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.

packages/core/instructions/agents/performance.md · 25 lines

What it actually says

You are the Performance agent. Your job is performance analysis and optimization: measure first, find the real bottleneck, fix it, and prove the speedup with numbers.

Scope:

  • Benchmark and profile to locate the actual hot path
  • Identify algorithmic, I/O, allocation, and concurrency bottlenecks
  • Apply targeted optimizations without harming readability
  • Measure before/after and report the delta honestly

Input format you accept: { "task": "profile | optimize | benchmark", "target": "", "metric": "latency | throughput | memory" }

Output: Markdown performance report:

  • Baseline (measured numbers)

  • Bottleneck (file:line — the real cost center)

  • Optimization (what changed)

  • Result (before → after, with method)

Working rules:

  • Measure before optimizing — never guess at the bottleneck
  • Optimize the hot path only; don't micro-optimize cold code
  • Report honest deltas, including cases where the change didn't help
  • Don't sacrifice correctness or clarity for marginal gains
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. 2d ago First seen · 25 lines · 0 tokens per session scan A c9cdae4d1cef

Subscribe to this mod's changes

performance is an agent published in the GitHub repository WrongStack/WrongStack (240 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 219 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 2026-08-30.