no-jumping-ahead

no-jumping-ahead is a cursor rule for Cursor from Brad-Edwards/aptl. It costs 407 tokens per session, scanned A, original, MIT.

A set of rules for making only the changes explicitly requested in a codebase. It also says to check unclear requests before taking extra action.

In plain words
What is it for?
Keeping implementation tasks focused, deciding when to ask for clarification, and stopping once the requested work is done.
Why use it?
It reduces unwanted files, features, setup work, and edits outside the requested scope.

Cursor rule for Cursor

Written for Cursor: installed under .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/brad-edwards/aptl/no-jumping-ahead
Clone the repo
git clone --depth 1 https://github.com/Brad-Edwards/aptl

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.

agentmods badge for no-jumping-ahead

README.md
[![agentmods](https://agentmods.dev/badge/rules/brad-edwards/aptl/no-jumping-ahead.svg)](https://agentmods.dev/rules/brad-edwards/aptl/no-jumping-ahead)
Your own site
<a href="https://agentmods.dev/rules/brad-edwards/aptl/no-jumping-ahead"><img src="https://agentmods.dev/badge/rules/brad-edwards/aptl/no-jumping-ahead.svg" alt="Measured on agentmods" height="20"></a>
Per session 407 This file is loaded in full into every session.
When invoked 407 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 5.1 $0.00407 $0.00407
Opus 5 $0.00204 $0.00204
Sonnet 5 $0.00081 $0.00081
Haiku 4.5 $0.00041 $0.00041

Measured 6d ago against content hash 1ab26ee8ec54, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

no-jumping-ahead 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 6d 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/no-jumping-ahead.mdc · 51 lines

What it actually says

DO NOT JUMP AHEAD

ALWAYS follow these rules when working on this codebase:

  1. ONLY do what is explicitly requested - Do not add extra features, scripts, or files that weren't asked for
  2. DO NOT assume what the user wants - If unclear, ask for clarification rather than making assumptions
  3. CHECK before going beyond the request - If you think something additional might be helpful, ask first
  4. STOP when the task is complete - Don't continue adding "helpful" extras
  5. RESPECT the user's workflow - They know their codebase and requirements better than you do

What NOT to do:

  • Create helper scripts without being asked
  • Add complex setup procedures when simple ones exist
  • Generate documentation for features that don't exist yet
  • Modify files beyond what was specifically requested
  • Add "best practices" that weren't requested

What TO do:

  • Answer the specific question asked
  • Make only the changes explicitly requested
  • Ask for clarification when requirements are unclear
  • Keep responses focused and concise
  • Respect existing code patterns and structure

Remember: Being helpful means doing exactly what was asked, not what you think might be helpful.

WRITING STYLE GUIDELINES

ALWAYS follow these writing rules for this technical/cyber audience:

DO NOT:

  • Use marketing language, boosterism, or sales-y content
  • Write verbose explanations or "value propositions"
  • Use excessive emojis or excitement
  • Create bulleted lists with bold formatting everywhere
  • Write like you're selling something

DO:

  • Use plain, simple, direct language
  • Write for busy technical professionals
  • Be concise and to the point
  • Focus on facts and functionality
  • Remember: technical people get turned off by marketing speak

Remember: This audience wants information, not persuasion. Write documentation like you're talking to a colleague, not a prospect.

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. 6d ago First seen · 51 lines · 407 tokens per session scan A 1ab26ee8ec54

Subscribe to this mod's changes

no-jumping-ahead is a cursor rule published in the GitHub repository Brad-Edwards/aptl (24 stars, last pushed yesterday), licensed MIT. It adds 407 tokens to every session, about $0.0020 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 2026-08-30.