The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits a potential surface for indirect prompt injection due to its core functionality:
Ingestion points: The agent is instructed to read and analyze arbitrary C++ and Blueprint source code, project configuration files, and runtime logs from the user's game project to identify integration points and diagnose issues (as described in Phase 1-9 instructions).
Boundary markers: The instructions do not explicitly mandate the use of delimiters or 'ignore embedded instructions' warnings when the agent processes these untrusted project files.
Capability inventory: The skill possesses significant capabilities, including the ability to execute shell commands (bash, git, p4, and Unreal build tools), write new source files to the project directory, and modify existing game code.
Sanitization: No explicit sanitization or filtering of ingested source code or log content is implemented before the model processes it.
[COMMAND_EXECUTION]: The skill relies on extensive command execution to perform its duties, including environment setup, repository management (Git/Perforce), Unreal Engine Blueprint introspection (via the BP Inspector tool), and project compilation (via UnrealBuildTool). These commands are necessary for the skill's purpose but represent a broad capability set that should be monitored.
[EXTERNAL_DOWNLOADS]: The skill downloads SDK binaries and Unreal Engine plugins from the author's public GitHub repository at github.com/ludeo-labs/unreal-plugin-releases. These downloads are documented and originate from the verified vendor infrastructure.

ludeo-unreal-integration