Deep UE5 Blueprint Audit

VerifiedCaution

Comprehensive audit of exported Blueprints: pattern scanning, graph health checks, and cross-function bug analysis.

Sby Skills Guide Bot
DevelopmentIntermediate
10/22/2026
Claude Code
#unreal-engine#blueprint#static-analysis#code-audit#ue5

Recommended for

Automated summary

Performs a deep audit of previously exported Unreal Engine Blueprint files (T3D or snapshots), detecting known bug patterns, graph health problems and cross-function conflicts.

Strengths

  • Runs patterns P-001 through P-025 on every function and labels each match CERTAIN, PROBABLE or POSSIBLE.
  • Ships a concrete graph-health checklist: orphaned nodes, unguarded casts, unwired ForEachLoop Completed pins, missing Switch/Select cases.
  • Adds cross-function analysis for shared variables, Save/Restore field-index mismatches and parameter propagation.
  • Emits a fixed report template with execution traces, suggested fixes and a final verdict.

Limitations

  • Requires a prior blueprint-export run; Blueprints that were never exported cannot be analyzed.
  • Depends on external knowledge-base files (ue5-universal-patterns.md, language-rules.md) existing at the referenced paths.
  • Heuristic findings are explicitly probabilistic and may flag intentional design choices as POSSIBLE issues.
When to use it

Use it right after exporting Blueprint data, when you want a thorough bug and graph-health review before refactoring or a code review.

When not to use it

Avoid it before running blueprint-export, or for reviewing C++ code or non-Blueprint assets.

Security analysis

Caution
Quality score78/100

The skill is a read/analysis workflow for exported Unreal Blueprint files and contains no destructive, network, or secret-handling instructions, so it poses no direct harm. It is rated caution rather than safe only because it grants unrestricted Bash and Write despite never needing them, giving an executing agent more capability than the documented steps require.

Findings
  • •Declared allowed-tools include Bash, but no step in the workflow actually specifies a shell command; the task is purely file reading/analysis, so Bash is an unnecessary privilege grant that lets an agent run arbitrary commands under this skill.
  • •Write is declared, but the report step only shows a text template with no defined output path; an agent could create or overwrite files in the project without guidance.
  • •No prompt-injection, obfuscation, exfiltration, or destructive command patterns found; external file reads are limited to plugin data and knowledge-base paths.
  • •Bash is declared in allowed-tools but never invoked or referenced anywhere in the skill body; Write is likewise declared but the skill only describes presenting a text report — declared capabilities exceed actual need (minor least-privilege concern, not an exploit).
  • •Skill reads files from ${CLAUDE_SKILL_DIR}/../../data/ and ${CLAUDE_PLUGIN_DATA}/ — path traversal outside the skill directory is read-only and consistent with a plugin data layout, but the referenced knowledge-base files are not bundled/verifiable here.
How this analysis is made

Examples

Full audit after export
/ue5-blueprints:blueprint-audit — audit all the T3D and snapshot files in the plugin data folder and report every issue by severity.
Graph health check
Audit the exported Blueprint functions for orphaned nodes, unguarded casts and unwired ForEachLoop Completed pins, then give me the verdict.
Cross-function conflicts
Check whether the exported functions share variables with write conflicts or have mismatched Save/Restore field indices, and show me the cross-function section of the report.

name: blueprint-audit description: Deep audit of exported Blueprint data — full pattern scan, graph health checks, and cross-function analysis. Use after blueprint-export has generated T3D or snapshot files. allowed-tools: [Read, Write, Glob, Grep, Bash]

Blueprint Audit

Perform a comprehensive audit of previously exported Blueprint data. This is the deep analysis — it reads exported files and runs every pattern and heuristic check.

Before Starting

Read the knowledge base:

  1. ${CLAUDE_SKILL_DIR}/../../data/patterns/ue5-universal-patterns.md
  2. ${CLAUDE_SKILL_DIR}/../../data/language-rules.md

Steps

1. Find Export Data

Look for exported files in ${CLAUDE_PLUGIN_DATA}/:

  • .t3d files (T3D exports)
  • *_snapshot.json files (graph snapshots)

If no exports found, tell the user to run /ue5-blueprints:blueprint-export first.

2. Parse and Map

For each Blueprint function/graph:

  • List all nodes with friendly names
  • Trace execution flow (entry → ... → return)
  • Map data wiring (which pins connect to which)
  • Identify all variables (Get and Set)
  • Note all default/literal values

3. Pattern Scan

Run every pattern P-001 through P-025 against each function. For each match:

Tag with confidence:

  • CERTAIN — You traced the wires and confirmed the bug
  • PROBABLE — Strong signal but needs context
  • POSSIBLE — Heuristic flag, might be intentional

4. Graph Health Checks

For each function:

  • [ ] Dead-end execution paths?
  • [ ] Orphaned nodes (no connections)?
  • [ ] Cast without failure wiring?
  • [ ] ForEachLoop with unwired Completed?
  • [ ] Array Get without bounds check?
  • [ ] Write-only variables (Set but never Get)?
  • [ ] Read-before-write variables?
  • [ ] Select/Switch with missing cases?

5. Cross-Function Analysis

If multiple functions were exported:

  • Do any functions share variables? Check for write conflicts.
  • Are there Save/Restore pairs? Check for field index mismatches.
  • Does the call chain propagate required parameters (like actor references)?

6. Report

=== AUDIT: [BlueprintName] ===

FUNCTIONS SCANNED: [count]
ISSUES FOUND: [count by severity]

--- [FunctionName] ---

EXEC FLOW:
  Entry → Node1 → Node2 → ... → Return

BUGS:
  [Bug1] P-001 CRITICAL (CERTAIN) — BreakStruct copy on Array_Set_1
    Plain English: [explanation]
    Fix: [steps using language rules]

HEURISTIC FINDINGS:
  [H1] ForEachLoop_1.Completed disconnected (POSSIBLE)
  [OK] No orphaned nodes
  [OK] No write-only variables

CROSS-FUNCTION:
  [Any dependency issues]

VERDICT: [Clean / Needs Fixes / Critical Issues]

Present CERTAIN findings first, then PROBABLE, then POSSIBLE.

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/deep-ue5-blueprint-audit-mvda87p1/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/deep-ue5-blueprint-audit-mvda87p1' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/deep-ue5-blueprint-audit-mvda87p1/download' --output '.skill-downloads/deep-ue5-blueprint-audit-mvda87p1/SKILL.md.part' && mv '.skill-downloads/deep-ue5-blueprint-audit-mvda87p1/SKILL.md.part' '.skill-downloads/deep-ue5-blueprint-audit-mvda87p1/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/deep-ue5-blueprint-audit-mvda87p1/SKILL.md' && test -s '.skill-downloads/deep-ue5-blueprint-audit-mvda87p1/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/blueprint-audit' && cp -R '.skill-downloads/deep-ue5-blueprint-audit-mvda87p1' '.claude/skills/blueprint-audit'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/deep-ue5-blueprint-audit-mvda87p1'

openskills read 'deep-ue5-blueprint-audit-mvda87p1'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/deep-ue5-blueprint-audit-mvda87p1/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills