loomle AGENTS.md

Repository instructions for Loomle, a project that gives coding agents tools for operating Unreal Engine, the software used to build games and 3D applications. They define how Loomle should reflect Unreal Engine’s own behavior and how its public tools should be named.

In plain words
What is it for?
Developing or reviewing Loomle features that interact with Unreal Engine. They guide source-code research, public MCP tool naming, internal bridge names, and readable protocol values.
Why use it?
They help contributors avoid inventing behavior that differs from Unreal Engine and prevent confusion between public tools and internal communication methods. They also make protocol names consistent.

Instructions file for CodexOpenCode

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 instructions/loomle/loomle/agents-md
Clone the repo
git clone --depth 1 https://github.com/loomle/loomle

Made for: Codex, OpenCode.

Per session 877 This file is loaded in full into every session.
When invoked 877 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 $0.00877 $0.00877
Opus 5 $0.00439 $0.00439
Sonnet 5 $0.00175 $0.00175
Haiku 4.5 $0.00088 $0.00088

Measured 2d ago against content hash 8d13a8bbe24f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

loomle AGENTS.md 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.

AGENTS.md · 103 lines

How it starts

The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Loomle Working Rules

Goal

Loomle exists to help agents operate Unreal Engine smoothly, reliably, and in terms that match UE semantics.

Loomle should not invent a replacement model for UE, and its progress should not depend on continually discovering new product ideas. Its work is to compare agent workflows against UE itself, find the places where direct operation is awkward, fragile, or unclear, and expose those capabilities through clear, faithful tools.

UE source code is Loomle's most important context. Any change that touches UE behavior, interface semantics, or tool boundaries must start by reading the relevant UE source and then designing Loomle's expression of that behavior.

Public Protocol Naming

Public MCP tool names use readable snake_case: status, project, sal_query, sal_patch, sal_schema, agent_skill, and editor. Private Bridge RPC tool names use dotted transport names such as sal.query, editor.context, editor.open, and editor.close; they are not public MCP tools. Public enum-like protocol values, including view, operation, kind, reason, and error code fragments, use readable snake_case, such as exec_flow, data_flow, and open_exec_output.

JSON field names should remain consistent with the existing public surface. Unreal C++ implementation code should continue to follow Unreal naming conventions and UE API names.

Mutation Dry Run Contract

Tools that mutate UE state must follow docs/MUTATION_DRY_RUN_CONTRACT.md. In short, dryRun=true should share the real edit path through parse, resolve, validate, and plan, then stop before applying changes. Do not expose expectedRevision, returnDiff, or rich dry-run result fields for a tool until that tool actually enforces or returns them. Bridge-side mutation result fields should be assembled through the shared LoomleMutation utility in engine/LoomleBridge/Source/LoomleBridge/Private/LoomleMutationResult.h, with tool modules supplying only their UE-specific parse, resolve, validate, plan, and apply logic.

Read the full file on GitHub · 103 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 · 103 lines · 877 tokens per session scan A 8d13a8bbe24f

Subscribe to this mod's changes

loomle AGENTS.md is an instructions file published in the GitHub repository loomle/loomle (13 stars, last pushed 6d ago), licensed MIT. It adds 877 tokens to every session, about $0.0044 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.