AI Agent Hub
Back to plugins
🖥️

dsh-UEAssetsOperator

Client Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install QSWWLTN/dsh-UEAssetsOperator

Paste the following prompt into your AI chat to install this plugin:

Run the dsh plugin install command shown above in your DeepSeek Harness terminal to install the plugin from https://github.com/QSWWLTN/dsh-UEAssetsOperator into your current profile.

About this plugin

In an AI-assisted Unreal Engine workflow, inspecting the type, tags, dependencies, and references of a .uasset file, or tweaking a Blueprint node connection, often costs more time than the actual feature work. A general-purpose chat model can only guess at binary content through strings or hex dumps, which is both unreliable and slow. dsh-UEAssetsOperator lets the AI inside DeepSeek Harness call native tools directly for both tasks, eliminating the need to hand-roll parsing scripts.

The plugin registers two native tools: ue_uasset_inspect and ue_blueprint_python_edit. The inspect tool offers three modes: resolve parses only the .uproject and virtual package paths; registry (the default) reads type, tags, dependencies, and references through the Asset Registry; load actually loads the UObject and extracts selected properties and metadata. Blueprint editing runs a restricted Python script that touches only the specified nodes rather than the entire graph. All path resolution starts from the session working directory or the nearest .uproject root and performs a bounded search; a full-disk scan is never attempted, and a missing path is reported instead of widening the scope. When the user message contains intent keywords such as .uasset, /Game/..., Blueprint, or DataTable, the plugin proactively injects the tool schemas and a full operation guide into the current turn so the model can invoke them without extra prompting.

This plugin is well suited for engineers who maintain UE projects or plugins inside a DeepSeek Harness environment: bulk-checking asset metadata, tracing broken dependency chains, fine-tuning Blueprint node connections with scripts, or simply wanting the AI to replace the manual round-trip of opening the editor to read values one by one. There are no extra npm runtime dependencies; the Cordis runner and the tool and skill registries are supplied by the host, so the plugin is ready to use out of the box.

Use Cases

  • Batch-checking asset type, tags, dependencies, and references
  • Tweaking Blueprint node connections via a restricted script without opening the editor
  • Resolving asset metadata under /Game/... paths directly in an AI chat

Best For

  • Engineers maintaining UE projects or plugins inside a DSH environment
  • QA or tool developers who need to batch-verify asset metadata and dependency chains
  • Developers who want AI to replace manual editor round-trips for value inspection