execution-guard

A project-protection rule that tells the coding agent to stop when a requested change conflicts with the project's architecture, maintainability standards, or practical value.

In plain words
What is it for?
It helps review requests against architecture documents, data-driven design, modularity, multiplayer requirements, and editor-friendly patterns before execution.
Why use it?
It prevents changes that could break established design, create hard-to-maintain code, or undermine the project's structure.

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/execution-guard
Clone the repo
git clone --depth 1 https://github.com/Stramp/Lost-Mines-Phandelver

Made for: Cursor.

Per session 2,225 This file is loaded in full into every session.
When invoked 2,225 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 $0.02225 $0.02225
Opus 5 $0.01112 $0.01112
Sonnet 5 $0.00445 $0.00445
Haiku 4.5 $0.00222 $0.00222

Measured 2d ago against content hash f503828a0821, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

execution-guard 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/execution-guard.mdc · 310 lines

How it starts

The opening of the file, as written. The whole thing — 310 lines — stays where its author put it; the contents beside it link to each section on GitHub.

🛡️ Execution Guard - Project Protection Rule

🚨 CRITICAL RULE - MAXIMUM PRIORITY 🚨

NEVER execute requests that:

  • ❌ Conflict with project architecture (see ARCHITECTURE.md)
  • ❌ Violate Clean Code or Design Patterns
  • ❌ Are unviable or overly complex for the benefit
  • ❌ Would create code that's hard to maintain
  • ❌ Would break established patterns

ALWAYS stop execution and report when detecting these issues.

When to STOP Execution

1. Architecture Conflicts

STOP if request:

  • Violates Data-Driven principle (hardcodes rules instead of using Data Tables/Assets)
  • Breaks Component-Based Architecture (mixes layer responsibilities)
  • Conflicts with Multiplayer-Ready design (missing replication, wrong authority)
  • Doesn't follow Editor-Friendly patterns
  • Breaks modularity (creates circular dependencies, wrong directory structure)

Example:

User: "Create a function that hardcodes race bonuses directly in the component"

Agent: [STOPS] "I cannot execute this request. It conflicts with our
         Data-Driven architecture principle. According to ARCHITECTURE.md,
         all rules must come from Data Tables or Data Assets, never hardcoded.

         This would:
         - Violate Data-Driven principle
         - Make it impossible for designers to modify rules
         - Break our established architecture

         Correct approach: Use UDataTable for race bonuses and load them
         at runtime. This aligns with our architecture and maintains
         flexibility for designers."

2. Clean Code Violations

STOP if request would:

  • Create helper functions inside Motor/Validator/Updater files (must be in Helpers/)
  • Create "local helper" functions that could be reused elsewhere
  • Create functions > 50 lines without refactoring
  • Create deeply nested conditionals (> 3-4 levels)
  • Duplicate code instead of using helpers
  • Create helper functions inside classes when they should be in Utils/
  • Create "temporary" or "just for testing" code that violates patterns
  • Create files with hundreds of lines of helpers

Read the full file on GitHub · 310 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 · 310 lines · 2,225 tokens per session scan A f503828a0821

Subscribe to this mod's changes

execution-guard is a cursor rule published in the GitHub repository Stramp/Lost-Mines-Phandelver (2 stars, last pushed 9mo ago), licensed MIT. It adds 2,225 tokens to every session, about $0.0111 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.