self-improve

Guidelines for updating Cursor rules as repeated code patterns, errors, tools, and practices appear in a codebase.

In plain words
What is it for?
Use it to spot patterns across files, decide when rules need adding or changing, and improve examples and guidance.
Why use it?
It helps keep project rules aligned with how the code is actually written and with recurring review or error patterns.

Cursor rule for Cursor

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 rules/nirarad/playwright-ai-qa-agent/self-improve
Clone the repo
git clone --depth 1 https://github.com/nirarad/playwright-ai-qa-agent

Made for: Cursor.

Per session 2,736 This file is loaded in full into every session.
When invoked 2,736 The same file — it is already loaded in full.
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 $0.02736 $0.02736
Opus 5 $0.01368 $0.01368
Sonnet 5 $0.00547 $0.00547
Haiku 4.5 $0.00274 $0.00274

Measured 2d ago against content hash f7b027322544, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

self-improve 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.

.cursor/rules/self-improve.mdc · 399 lines

How it starts

The opening of the file, as written. The whole thing — 399 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Rule Improvement Triggers

  • New code patterns not covered by existing rules
  • Repeated similar implementations across files
  • Common error patterns that could be prevented
  • New libraries or tools being used consistently
  • Emerging best practices in the codebase

Analysis Process:

  • Compare new code with existing rules
  • Identify patterns that should be standardized
  • Look for references to external documentation
  • Check for consistent error handling patterns
  • Monitor test patterns and coverage

Rule Updates:

  • Add New Rules When:

    • A new technology/pattern is used in 3+ files
    • Common bugs could be prevented by a rule
    • Code reviews repeatedly mention the same feedback
    • New security or performance patterns emerge
  • Modify Existing Rules When:

    • Better examples exist in the codebase
    • Additional edge cases are discovered
    • Related rules have been updated
    • Implementation details have changed
  • Example Pattern Recognition:

    // If you see repeated patterns like:
    const data = await prisma.user.findMany({
      select: { id: true, email: true },
      where: { status: 'ACTIVE' }
    });
    
    // Consider adding to [prisma.mdc](mdc:shipixen/.cursor/rules/prisma.mdc):
    // - Standard select fields
    // - Common where conditions
    // - Performance optimization patterns
    
  • Rule Quality Checks:

  • Rules should be actionable and specific

  • Examples should come from actual code

  • References should be up to date

  • Patterns should be consistently enforced

Continuous Improvement:

  • Monitor code review comments
  • Track common development questions
  • Update rules after major refactors
  • Add links to relevant documentation
  • Cross-reference related rules

Rule Deprecation

  • Mark outdated patterns as deprecated
  • Remove rules that no longer apply
  • Update references to deprecated rules
  • Document migration paths for old patterns

Documentation Updates:

  • Keep examples synchronized with code
  • Update references to external docs
  • Maintain links between related rules
  • Document breaking changes

Read the full file on GitHub · 399 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. 2d ago First seen · 399 lines · 2,736 tokens per session scan A f7b027322544

Subscribe to this mod's changes

self-improve is a cursor rule published in the GitHub repository nirarad/playwright-ai-qa-agent (5 stars, last pushed 4mo ago), licensed MIT. It adds 2,736 tokens to every session, about $0.0137 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.