agentleFS
PeterTXPanPeterTXPan/dsh-unreal-mcp/skills/unreal-skill/SKILL.md

unreal-skill

Skill16 starsChanged 3 months agoMITView on GitHub

Use when creating, editing, or reviewing an Unreal Engine Agent Skill registered inside the Editor as a Python UAgentSkill or UAsset. This is distinct from a DSH harness Skill; skip Toolset authoring and ordinary use of an existing Unreal Skill.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. EpicGames/unreal-engine-skills-for-claude-code-plugin★ 1.8ksame name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 7 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Durable: avoid volatile property and tool names when discovery can provide them at runtime.

    line 22
  2. Distinctive:

    Harness-agnostic: do not mention DSH, Claude Code, model names, or agent wiring inside the UE Skill payload.

    line 24
  3. Distinctive:

    Use Unreal MCP Tool Search to discover the Agent Skill Toolset and its current schema.

    line 30
  4. Very specific:

    Follow the plugin's existing import path: Skill modules normally live under a skills package imported by init_unreal.py.

    line 73
  5. Distinctive:

    Use the current Agent Skill Toolset schema discovered through MCP.

    line 81
  6. Distinctive:

    Never guess a path or repeat a timed-out create/update call: first query the Editor to determine whether the intended Skill already exists …

    line 84
  7. Distinctive:

    no volatile tool catalog is embedded.

    line 97
  8. Distinctive:

    only explicitly authorized files/assets changed and intended packages were saved.

    line 99

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Unreal Skill

Author or review an Unreal Engine Agent Skill: reusable instructions registered inside the UE Editor. This is not a DSH Skill stored in this Bundle. Use create-toolset for new callable Toolsets and unreal-mcp when merely invoking an existing UE Skill or Editor tool.

Principles

A useful UE Agent Skill is:

  • Novel: teach durable knowledge the agent cannot discover from tool schemas or direct state inspection.
  • Collegial: brief a knowledgeable UE colleague rather than reproducing a manual.
  • Flexible: prescribe order only where UE correctness or safety requires it.
  • Durable: avoid volatile property and tool names when discovery can provide them at runtime.
  • Harness-agnostic: do not mention DSH, Claude Code, model names, or agent wiring inside the UE Skill payload.
  • Parsimonious: justify every token added to the agent context.

Before writing

  1. Use Unreal MCP Tool Search to discover the Agent Skill Toolset and its current schema. List registered UE Skills and inspect relevant ones before creating a duplicate.
  2. Confirm the exact project/plugin and obtain authorization before creating or modifying Python files, Content Browser assets, plugin registration, or Editor settings.
  3. Choose the implementation based on ownership:
    • Python UAgentSkill: code-plugin guidance that belongs in version control and loads with the plugin.
    • UAsset Skill: project-specific guidance stored in Content and managed through the live Editor.

Skill contract

Each UE Agent Skill has a concise discovery description and an instruction payload. The description must let an agent decide when the Skill applies without loading the body. Instructions should contain only non-obvious, reusable guidance and assume the agent will discover current Toolsets and schemas at runtime.

Python Skills

Define a subclass of unreal.AgentSkill, decorate it with the agent-skill decorator supplied by ToolsetRegistry, use the class docstring as the discovery description, and place the instruction payload in its instructions class attribute.

import unreal
from toolset_registry.agent_skill import agent_skill

_INSTRUCTIONS = (
    "Do X before Y because Z.\n"
    "Verify the resulting Editor state after the operation.\n"
)

@agent_skill
class MySkill(unreal.AgentSkill):
    """Guidance for X workflows in Unreal Engine."""

    instructions = _INSTRUCTIONS

Follow the plugin's existing import path: Skill modules normally live under a skills package imported by init_unreal.py. They register on import. After an authorized edit, reload the owning Python package before verification. Enabling Python Remote Execution is a separate security-sensitive setting and must not be changed silently.

UAsset Skills

Use the current Agent Skill Toolset schema discovered through MCP. For creation, provide an authorized Content folder, PascalCase asset name, short description, and instructions. For updates, target the exact existing Skill object path. Never guess a path or repeat a timed-out create/update call: first query the Editor to determine whether the intended Skill already exists and what content it contains.

Review and verification

Query the saved or registered Skill through the Agent Skill Toolset, then read its description and instructions together as the agent will receive them. Verify:

  • the description routes only relevant tasks;
  • instructions add knowledge unavailable from tools;
  • the payload contains no harness-specific assumptions;
  • no volatile tool catalog is embedded;
  • the correct Python class or UAsset path is registered;
  • only explicitly authorized files/assets changed and intended packages were saved.

Report the implementation path, registered name/path, reload or save result, verification query, and any changes the user still needs to persist.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.