ask-mode-behavior

ask-mode-behavior is a cursor rule for Cursor from Stramp/Lost-Mines-Phandelver. It costs 568 tokens per session, scanned A, original, MIT.

A response rule for keeping answers in an ask-only mode extremely brief. It also forbids code blocks, file creation, and detailed implementation advice in that mode.

In plain words
What is it for?
Use it when answering questions in ask mode, especially when a one-sentence explanation is enough.
Why use it?
It prevents overly long responses and keeps read-only questions focused on explanations rather than changes.

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/stramp/lost-mines-phandelver/ask-mode-behavior
Clone the repo
git clone --depth 1 https://github.com/Stramp/Lost-Mines-Phandelver

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 ask-mode-behavior

README.md
[![agentmods](https://agentmods.dev/badge/rules/stramp/lost-mines-phandelver/ask-mode-behavior.svg)](https://agentmods.dev/rules/stramp/lost-mines-phandelver/ask-mode-behavior)
Your own site
<a href="https://agentmods.dev/rules/stramp/lost-mines-phandelver/ask-mode-behavior"><img src="https://agentmods.dev/badge/rules/stramp/lost-mines-phandelver/ask-mode-behavior.svg" alt="Measured on agentmods" height="20"></a>
Per session 568 This file is loaded in full into every session.
When invoked 568 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 $0.00568 $0.00568
Opus 5 $0.00284 $0.00284
Sonnet 5 $0.00114 $0.00114
Haiku 4.5 $0.00057 $0.00057

Measured 3d ago against content hash 1aa97807a6c2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ask-mode-behavior 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 3d 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/ask-mode-behavior.mdc · 79 lines

How it starts

The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.

🚨 Behavior in ASK Mode - CRITICAL RULE 🚨

⚠️ MANDATORY - MAXIMUM PRIORITY ⚠️

THIS RULE HAS PRIORITY 1 AND MUST BE FOLLOWED IN ALL ASK MODE SESSIONS.

BEFORE responding in ASK Mode, ALWAYS verify:

  • Am I in ASK Mode? (read-only tools available)
  • Is my response ONE sentence or extremely brief?
  • Am I NOT showing code blocks?
  • Am I NOT creating files?
  • Am I being direct and to the point?

If ANY answer is NO → STOP and rewrite response to be brief.

Critical Rules for ASK Mode

When the user is in ASK Mode:

  1. NEVER write code

    • Do not show complete code blocks
    • Do not create files
    • Do not suggest detailed implementations
    • Only explain concepts or direct to Agent Mode
  2. ALWAYS respond extremely briefly

    • ONE sentence when possible
    • Straight to the point
    • NO long explanations
    • NO extensive examples
    • NO detailed reports
    • NO multi-paragraph responses
  3. Exception: Complete Explanations

    • Only when the user EXPLICITLY asks for "complete explanation" or similar
    • Otherwise, keep response in one sentence

Response Examples

❌ WRONG (too long):

To create a Data Asset you need to first create a class that inherits from UDataAsset, then add properties with UPROPERTY, and then...

[complete code here]

✅ CORRECT (brief):

Create a class that inherits from UDataAsset with UPROPERTY(EditDefaultsOnly). Switch to Agent Mode for implementation.

When asking for complete explanation:

User: "Explain completely how Data Assets work"

Response: [Here can be detailed, with examples and explanations]

Direction to Agent Mode

If the user asks for something that requires code:

  • Respond briefly
  • Direct to Agent Mode: "Switch to Agent Mode (Ctrl+L) to implement"

🚨 Enforcement

This rule is MANDATORY and has priority 1. Violations of this rule are CRITICAL errors. Always verify compliance before responding in ASK Mode.

Read the full file on GitHub · 79 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. 3d ago First seen · 79 lines · 568 tokens per session scan A 1aa97807a6c2

Subscribe to this mod's changes

ask-mode-behavior is a cursor rule published in the GitHub repository Stramp/Lost-Mines-Phandelver (2 stars, last pushed 9mo ago), licensed MIT. It adds 568 tokens to every session, about $0.0028 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.