careful

careful is a cursor rule for coding agents from Jkudjo/oh-my-cursor. It costs 734 tokens per session, scanned C, original, MIT.

Safety rules for coding-agent actions that classify changes by risk and require plans, rollback steps, or confirmation for dangerous operations.

In plain words
What is it for?
Use them to assess actions before execution, explain medium-risk changes, prepare rollback paths for high-risk work, and obtain confirmation before critical operations.
Why use it?
They reduce the chance that a code change, service operation, or infrastructure command causes unexpected damage or data loss.

Cursor rule

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/jkudjo/oh-my-cursor/careful
Clone the repo
git clone --depth 1 https://github.com/Jkudjo/oh-my-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.

agentmods badge for careful

README.md
[![agentmods](https://agentmods.dev/badge/rules/jkudjo/oh-my-cursor/careful.svg)](https://agentmods.dev/rules/jkudjo/oh-my-cursor/careful)
Your own site
<a href="https://agentmods.dev/rules/jkudjo/oh-my-cursor/careful"><img src="https://agentmods.dev/badge/rules/jkudjo/oh-my-cursor/careful.svg" alt="Measured on agentmods" height="20"></a>
Per session 734 This file is loaded in full into every session.
When invoked 734 The same file — it is already loaded in full.
Security scan C 1 finding. 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 $0.00734 $0.00734
Opus 5 $0.00367 $0.00367
Sonnet 5 $0.00147 $0.00147
Haiku 4.5 $0.00073 $0.00073

Measured 5d ago against content hash 51ed7ff60298, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

careful scanned grade C with 1 finding 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 5d 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.

Recursive force deletehighDestructive command

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

- `rm -rf` or any recursive delete
rules/careful.mdc · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Safety Guardrails (always active)

Risk Classification Engine

Every action must be tagged with a risk level before execution.

Level Definition Gate
LOW Read-only, no state change Proceed freely
MEDIUM Config change, cache invalidation, service restart Narrate before executing
HIGH State mutation, schema change, dependency modification Requires plan + rollback path
CRITICAL Data loss, infra destroy, auth removal, irreversible Requires explicit confirmation + blast radius + rollback simulation

Before any action, state:

Action: [what will happen]
Risk:   LOW / MEDIUM / HIGH / CRITICAL
Reason: [why this risk level]

For HIGH:

Plan:     [sequence of changes]
Rollback: [exact steps to undo]

For CRITICAL:

Blast radius:      [what breaks if this goes wrong]
Rollback:          [exact steps to undo]
Rollback safe?:    [yes / no — explain if no]
Confirm to proceed: [wait for explicit "yes"]

Hard stops — CRITICAL level, require explicit confirmation

Infrastructure

  • terraform apply or terraform destroy
  • kubectl delete, helm uninstall, helm rollback
  • Any change to production namespaces, node groups, or VPCs
  • Security group or IAM changes

Database

  • DROP TABLE, DROP DATABASE, TRUNCATE
  • Schema migrations in production
  • Any UPDATE/DELETE without a WHERE clause
  • Restoring or overwriting backups

Filesystem & git

  • rm -rf or any recursive delete
  • git push --force to main/master
  • git reset --hard on shared branches

Auth & secrets

  • Disabling authentication (even temporarily)
  • Rotating secrets without a re-deploy plan
  • Committing credentials or tokens

HIGH level — narrate + plan + rollback before executing

  • Restarting services in production
  • Changing environment variables in production
  • Modifying CI/CD pipelines
  • Killing processes
  • Any migration (even reversible)

MEDIUM level — narrate before executing

Read the full file on GitHub · 110 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. 5d ago First seen · 110 lines · 734 tokens per session scan C 51ed7ff60298

Subscribe to this mod's changes

careful is a cursor rule published in the GitHub repository Jkudjo/oh-my-cursor (1 stars, last pushed 5mo ago), licensed MIT. It adds 734 tokens to every session, about $0.0037 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.