Claude
Skill
ue5-auto-assistant
UE5.6/UE5.7 automatic assistant entry for beginners. Use when users ask Unreal questions without naming a specific skill. Auto-route to the most precise UE5 skill and recommend dedicated MCP tools.
Virus-scanned
Reviewed automatically before listing.
Download
teixasalone-unrealengine5-skills-skills_ue5-auto-assistant-397623a.zip · 4 KB
Install
skills CLI
npx skills add https://github.com/teixasalone/UnrealEngine5-Skills/tree/main/skills/ue5-auto-assistant
Claude Code
claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install teixasalone-unrealengine5-skills@llmmart
Git
git clone https://github.com/teixasalone/UnrealEngine5-Skills.git
The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole teixasalone/unrealengine5-skills collection as a plugin from our marketplace. Git is the plain clone.
Skill manifest
Quick Start
- Treat this as the default entry for UE5.6/UE5.7 requests.
- Parse user intent first without requiring module names.
- Route to
ue5-module-routerwhen module-level precision is needed.
Workflow
- Detect request type: Blueprint, C++, UI, save/load, networking, world interaction, debugging, performance, packaging.
- If module names appear, delegate routing to
ue5-module-router. - If module names do not appear, route by intent:
- Blueprint ->
ue5-blueprint-workflow - C++ gameplay ->
ue5-cpp-gameplay - UI/UMG/Slate ->
ue5-ui-umg-slate - save/load/replication ->
ue5-save-load-replication - pickup/spawner/world interaction ->
ue5-world-interaction - perf/packaging ->
ue5-performance-packaging - debugging/validation ->
ue5-debug-validation - architecture/refactor ->
ue5-architecture
- Blueprint ->
- Return one primary skill and optional secondary skill for cross-domain requests.
- Return routing payload fields:
primary_skillsecondary_skillrecommended_mcp_tools[]route_confidenceroute_reason
Natural Language To Skill And Tools
- Blueprint requests:
- target skill:
ue5-blueprint-workflow - recommended tools:
blueprint_feature_build,blueprint_modify,blueprint_query
- target skill:
- C++ gameplay/system requests:
- target skill:
ue5-cpp-gameplay - recommended tools:
blueprint_query,asset_search,get_output_log
- target skill:
- UI/UMG/Slate requests:
- target skill:
ue5-ui-umg-slate - recommended tools:
blueprint_query,blueprint_modify,capture_viewport
- target skill:
- Save/load/replication requests:
- target skill:
ue5-save-load-replication - recommended tools:
character_data,blueprint_query,get_output_log
- target skill:
- World interaction/pickup/spawner requests:
- target skill:
ue5-world-interaction - recommended tools:
spawn_actor,get_level_actors,set_property,move_actor
- target skill:
- Performance/packaging requests:
- target skill:
ue5-performance-packaging - recommended tools:
run_console_command,get_output_log,capture_viewport,open_level
- target skill:
- Debug/validation requests:
- target skill:
ue5-debug-validation - recommended tools:
get_output_log,asset_search,blueprint_query,task_list
- target skill:
- Architecture/module-boundary requests:
- target skill:
ue5-architecture - recommended tools:
asset_search,asset_dependencies,asset_referencers
- target skill:
Constraints
- Do not require users to know skill names.
- Prefer deterministic routing with explicit reason.
- Keep fallback behavior explicit when confidence is low.
- Prefer dedicated MCP tools before
execute_script.
Failure Handling
- If intent is ambiguous, return top 2 route candidates and ask one short clarification.
- If request spans many systems, split into staged route steps.
- Clarification template:
Quick check: do you want A(<candidate_1>) or B(<candidate_2>)?- ask once, then continue.
Escalation
- Escalate when query depends on plugin/engine source outside indexed scope.
- Escalate when org-level coding standards are required but not available in repo.
Files (unrealengine5-skills)
-
agents
-
openai.yaml 304 B
display_name: UE5.6/UE5.7 Auto Assistant short_description: Beginner-friendly UE5.6/UE5.7 entry that auto-routes normal language requests to the right skill. default_prompt: Interpret this UE5.6/UE5.7 request from natural language and auto-route to the best skill, with one primary route and confidence.
-
-
references
-
beginner-smoke-prompts.md 1.7 KB
# Beginner Smoke Prompts (UE5.6/UE5.7 MCP + Skill) Use these prompts to validate natural-language routing and MCP tool priority. ## How To Use - Start a new Codex session. - Paste one case at a time. - Do not include skill names in prompts. - Verify expected routing and recommended tools. ## Case 1: Save/Load Prompt: `Inventory is empty after load. Where should I debug first?` Expected: - primary skill: `ue5-save-load-replication` - recommended tools include `get_output_log`, `blueprint_query` ## Case 2: Module-Specific Prompt: `How do I debug material compile stalls in RenderCore?` Expected: - routed by `ue5-module-router` - high confidence when exact module match exists ## Case 3: Blueprint Hotkey Prompt: `In Blueprint, press key 2 to jump and print a log. How do I wire this?` Expected: - primary skill: `ue5-blueprint-workflow` - fast-path for keyboard input (`add_input_key_event` first) ## Case 4: World Interaction Prompt: `I want pickup items that auto-collect when player enters range.` Expected: - primary skill: `ue5-world-interaction` - tools include `spawn_actor`, `get_level_actors`, `set_property` ## Case 5: UI Prompt: `My tooltip is clipped at viewport edges. How should I clamp it?` Expected: - primary skill: `ue5-ui-umg-slate` - tools include `blueprint_query`, `blueprint_modify`, `capture_viewport` ## Case 6: Performance/Packaging Prompt: `Give me a pre-package checklist and profiling order.` Expected: - primary skill: `ue5-performance-packaging` - tools include `run_console_command`, `get_output_log` ## Pass Criteria - correct routing for each prompt - dedicated MCP tools prioritized over `execute_script` - one clarification only when confidence is low -
mcp-skill-mapping.md 2.6 KB
# MCP Skill Mapping (UE5.6/UE5.7) This file defines skill-to-tool mapping for deterministic routing. ## ue5-auto-assistant - whitelist: - `status`, `get_ue_context`, `task_list`, `task_status` - graylist: - `get_output_log`, `capture_viewport` - avoid: - `execute_script` as first action ## ue5-module-router - whitelist: - `asset_search`, `asset_dependencies`, `asset_referencers`, `blueprint_query`, `get_output_log` - graylist: - `task_submit`, `task_status`, `task_result` - avoid: - direct world mutation before route confidence is established ## ue5-blueprint-workflow - whitelist: - `blueprint_feature_build`, `blueprint_modify`, `blueprint_query`, `enhanced_input` - graylist: - `asset_search`, `get_output_log` - avoid: - `execute_script` unless Blueprint tools cannot express the operation ## ue5-cpp-gameplay - whitelist: - `blueprint_query`, `asset_search`, `get_output_log`, `get_ue_context` - graylist: - `blueprint_modify` - avoid: - editor-world mutation tools for pure C++ design requests ## ue5-ui-umg-slate - whitelist: - `blueprint_query`, `blueprint_modify`, `asset_search`, `capture_viewport`, `get_output_log` - graylist: - `material`, `asset` - avoid: - `execute_script` for routine UMG/Slate wiring ## ue5-save-load-replication - whitelist: - `character_data`, `blueprint_query`, `blueprint_modify`, `asset`, `asset_search`, `get_output_log` - graylist: - `task_submit`, `task_status`, `task_result` - avoid: - client-side-only mutation assumptions without authority checks ## ue5-world-interaction - whitelist: - `spawn_actor`, `get_level_actors`, `set_property`, `move_actor`, `delete_actors`, `open_level`, `capture_viewport` - graylist: - `material`, `character` - avoid: - `run_console_command` for level switching when `open_level` exists ## ue5-debug-validation - whitelist: - `get_output_log`, `asset_search`, `blueprint_query`, `get_level_actors`, `capture_viewport`, `task_list`, `task_status`, `task_result` - graylist: - `run_console_command` - avoid: - broad state-changing operations during diagnosis ## ue5-performance-packaging - whitelist: - `run_console_command`, `get_output_log`, `capture_viewport`, `open_level`, `asset_dependencies`, `asset_referencers`, `asset_search` - graylist: - `task_submit`, `task_status`, `task_result` - avoid: - `execute_script` before standard profiling/validation tools are exhausted ## ue5-architecture - whitelist: - `asset_search`, `asset_dependencies`, `asset_referencers`, `blueprint_query`, `get_ue_context` - graylist: - `get_output_log` - avoid: - direct mutation tools unless explicitly requested -
natural-language-triggers.md 795 B
# Natural Language Triggers ## Example User Prompts - How do I add a keyboard hotkey in Blueprint? - Why is my inventory data empty after loading save data? - How do I write a reusable UE5 C++ component? - Tooltip goes off-screen in UMG. How should I fix it? - What should I check before packaging? ## Expected Routing - Blueprint behavior -> `ue5-blueprint-workflow` - C++ / class design -> `ue5-cpp-gameplay` - UI / UMG / Slate -> `ue5-ui-umg-slate` - save/load / replication -> `ue5-save-load-replication` - world interaction / pickup -> `ue5-world-interaction` - performance / packaging -> `ue5-performance-packaging` - debugging / validation -> `ue5-debug-validation` - module boundary / architecture -> `ue5-architecture` - if module names appear, route through `ue5-module-router` first
-
-
SKILL.md 3.2 KB
--- name: ue5-auto-assistant description: UE5.6/UE5.7 automatic assistant entry for beginners. Use when users ask Unreal questions without naming a specific skill. Auto-route to the most precise UE5 skill and recommend dedicated MCP tools. --- # Quick Start - Treat this as the default entry for UE5.6/UE5.7 requests. - Parse user intent first without requiring module names. - Route to `ue5-module-router` when module-level precision is needed. # Workflow - Detect request type: Blueprint, C++, UI, save/load, networking, world interaction, debugging, performance, packaging. - If module names appear, delegate routing to `ue5-module-router`. - If module names do not appear, route by intent: - Blueprint -> `ue5-blueprint-workflow` - C++ gameplay -> `ue5-cpp-gameplay` - UI/UMG/Slate -> `ue5-ui-umg-slate` - save/load/replication -> `ue5-save-load-replication` - pickup/spawner/world interaction -> `ue5-world-interaction` - perf/packaging -> `ue5-performance-packaging` - debugging/validation -> `ue5-debug-validation` - architecture/refactor -> `ue5-architecture` - Return one primary skill and optional secondary skill for cross-domain requests. - Return routing payload fields: - `primary_skill` - `secondary_skill` - `recommended_mcp_tools[]` - `route_confidence` - `route_reason` # Natural Language To Skill And Tools - Blueprint requests: - target skill: `ue5-blueprint-workflow` - recommended tools: `blueprint_feature_build`, `blueprint_modify`, `blueprint_query` - C++ gameplay/system requests: - target skill: `ue5-cpp-gameplay` - recommended tools: `blueprint_query`, `asset_search`, `get_output_log` - UI/UMG/Slate requests: - target skill: `ue5-ui-umg-slate` - recommended tools: `blueprint_query`, `blueprint_modify`, `capture_viewport` - Save/load/replication requests: - target skill: `ue5-save-load-replication` - recommended tools: `character_data`, `blueprint_query`, `get_output_log` - World interaction/pickup/spawner requests: - target skill: `ue5-world-interaction` - recommended tools: `spawn_actor`, `get_level_actors`, `set_property`, `move_actor` - Performance/packaging requests: - target skill: `ue5-performance-packaging` - recommended tools: `run_console_command`, `get_output_log`, `capture_viewport`, `open_level` - Debug/validation requests: - target skill: `ue5-debug-validation` - recommended tools: `get_output_log`, `asset_search`, `blueprint_query`, `task_list` - Architecture/module-boundary requests: - target skill: `ue5-architecture` - recommended tools: `asset_search`, `asset_dependencies`, `asset_referencers` # Constraints - Do not require users to know skill names. - Prefer deterministic routing with explicit reason. - Keep fallback behavior explicit when confidence is low. - Prefer dedicated MCP tools before `execute_script`. # Failure Handling - If intent is ambiguous, return top 2 route candidates and ask one short clarification. - If request spans many systems, split into staged route steps. - Clarification template: - `Quick check: do you want A(<candidate_1>) or B(<candidate_2>)?` - ask once, then continue. # Escalation - Escalate when query depends on plugin/engine source outside indexed scope. - Escalate when org-level coding standards are required but not available in repo.
Comments (0)
Sign in to join the conversation.
Reviews (0)
No reviews yet.
No comments yet.