Skip to main content
Glama

find_erroring_blueprints

Destructive

Locate Blueprints and AnimBlueprints stuck in compile error, reporting status, counts and the first error message, so you can fix the asset blocking Play-in-Editor.

Instructions

Scan a content path for Blueprints and AnimBlueprints sitting in a compile ERROR state and report each one with its status, error/warning counts and its FIRST error message -- the search for the culprit when PIE refuses to start because of an asset nobody was working on. It reads the loaded compile Status and recompiles the erroring ones with a results log to extract that message. WARNING: loading every Blueprint under a broad path is heavy enough to stall the editor on thousands of assets: scope path narrowly, and reach for only_loaded=true first -- the Blueprint that aborts PIE is usually still in memory, so that fast in-memory pass normally finds it without a single disk load. Complements validate_assets_in_path and audit_blueprint_health by listing EVERY erroring Blueprint with its first error in one call. WARNING: the scope parameter is named path, NOT search_path: pass the wrong name and it is silently ignored, the default /Game applies, and the tool sweeps the whole project while the answer looks scoped to your folder.

Returns: { path, scanned, inspected, only_loaded, erroring_count, warning_count, results_truncated, blueprints:[{ asset_path, name, class, status, num_errors, num_warnings, first_error }] }

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
pathNoContent path to scan, recursive (default "/Game"). WARNING: this exact name -- a `search_path` key is ignored and you get the full /Game sweep. Scope it narrowly to stay cheap.
forceNoOptional. This command is REFUSED while a Play-in-Editor session is running: it either compiles a Blueprint (the reinstancer can retain a world, and CheckForWorldGCLeaks is Fatal -- it KILLS the editor) or writes to disk (Editor.AllowSavingAssetsDuringPIE=0 makes the engine refuse the write, so it would report success without writing a byte). Stop PIE and retry instead; force=true runs it anyway, at your own risk.
max_scanNoMax Blueprints to load/inspect (default 5000).
asset_pathNoAlias of `path` (same value, same meaning). Accepted because sibling tools spell this parameter differently; prefer `path`.
compile_allNoRecompile EVERY scanned Blueprint (thorough but heavy); otherwise only recompile ones whose loaded Status is already Error. Default false.
max_resultsNoMax erroring/warning Blueprints to return (default 500).
only_loadedNoWhen true, only inspect Blueprints already loaded in memory (no disk load). Default false.
include_warningsNoAlso report Blueprints with compile WARNINGS (not just errors). Default false.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.1.0

TDQS

A4.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations declare destructive=true and non-readonly, and the description adds real behavioral context on top: it recompiles erroring assets with a results log, a broad-path scan can stall the editor on thousands of assets, and only_loaded avoids disk loads. These are cost and side-effect traits the annotations do not convey.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Dense and front-loaded: purpose first, then the fast-path advice, then sibling routing, then the parameter trap. It is long and repeats the path-naming warning that also appears in the schema, but nearly every sentence carries actionable content.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema, the description supplies the full return shape (path, scanned, inspected, erroring_count, blueprints[] with first_error). Combined with the PIE-refusal and scoping warnings, an agent has everything needed to call this costly, destructiveish scan safely.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3; the description goes beyond it by stating the path/only_loaded decision rule and the search_path trap, which is decision-level meaning rather than restated type info. It does not, however, explain max_scan/max_results semantics beyond what the schema already says.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb and resource (scan a content path for Blueprints/AnimBlueprints in a compile ERROR state) plus the exact diagnostic task it serves (finding the culprit when PIE won't start). It explicitly differentiates itself from validate_assets_in_path and audit_blueprint_health, so an agent can route correctly without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit when-to-use ordering ('reach for only_loaded=true first'), the scoping rule for `path`, and names the two sibling tools it complements. It even calls out a misuse trap (passing search_path is silently ignored and defaults to /Game), which is guidance no other field provides.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools