A set of guidance for choosing model tiers, managing limited conversation context, and handling complex work such as large refactors or difficult debugging. A context window is the amount of conversation and code an AI can consider at once.

It supports model selection, context planning, multi-file implementation, large refactors, complex debugging, and build troubleshooting.

It helps avoid losing important information or using an unsuitable model when a task is large or reasoning-heavy.

perf — Cursor rule for Cursor by timurgaleev/vibestack

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.

Clone the repo

git clone --depth 1 https://github.com/timurgaleev/vibestack

Made for: Cursor.

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.

README.md

[![agentmods](https://agentmods.dev/badge/rules/timurgaleev/vibestack/perf/github.svg)](https://agentmods.dev/rules/timurgaleev/vibestack/perf)

Your own site

<a href="https://agentmods.dev/rules/timurgaleev/vibestack/perf"><img src="https://agentmods.dev/badge/rules/timurgaleev/vibestack/perf/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/rules/timurgaleev/vibestack/perf"><img src="https://agentmods.dev/badge/rules/timurgaleev/vibestack/perf.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 329 This file is loaded in full into every session.

When invoked 329 The same file — it is already loaded in full.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00329	$0.00329
Opus 5	$0.00164	$0.00164
Sonnet 5	$0.00066	$0.00066
Haiku 4.5	$0.00033	$0.00033

Measured yesterday against content hash 6264dbd94d66, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-20, from the pricing page.

Security

Grade A, and why

perf 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.

config/cursor/rules/perf.mdc · 59 lines

What it actually says

Performance Optimization

Model Selection Strategy

Refer to model tiers, not pinned versions, so this guidance stays current as new releases ship. Always prefer the latest model in each tier.

Latest Opus tier (Deepest reasoning, PREFERRED):

Complex architectural decisions
All subagent workflows (task-planner, system-designer, etc.)
Main development work
Maximum reasoning requirements
Research and analysis tasks

Latest Sonnet tier (Best balance for coding):

Simple code generation when Opus is not needed
Quick edits and modifications
Orchestrating workflows

Latest Haiku tier (Fast and economical):

Simple utility functions
Documentation updates
Quick queries

Context Window Management

Avoid last 20% of context window for:

Large-scale refactoring
Feature implementation spanning multiple files
Debugging complex interactions

Lower context sensitivity tasks:

Single-file edits
Independent utility creation
Documentation updates
Simple bug fixes

For Complex Tasks

For complex tasks requiring deep reasoning:

Switch to Opus model for maximum reasoning
Plan the approach before writing any code
Break into independent subtasks
Use specialized subagents for isolated context and diverse analysis

Build Troubleshooting

If build fails:

Use build-doctor subagent
Analyze error messages
Fix incrementally
Verify after each fix

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.

yesterday First seen · 59 lines · 329 tokens per session scan A 6264dbd94d66

Subscribe to this mod's changes

perf is a cursor rule published in the GitHub repository timurgaleev/vibestack (6 stars, last pushed yesterday), licensed MIT. It adds 329 tokens to every session, about $0.0016 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 2026-09-19.

Building agents Claude Code agent-skills ai-agents cursor-ide developer-tools kiro

perf

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

Performance Optimization

Model Selection Strategy

Context Window Management

For Complex Tasks

Build Troubleshooting

What this file has done since we first saw it

Other cursor rules, from other repositories

tech-writer

ai-expert

back-end-dev

product-owner

devops

architect