cli-anything-godot
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions include a command to install the harness directly from the vendor's GitHub repository:
pip install git+https://github.com/HKUDS/CLI-Anything.git#subdirectory=godot/agent-harness. - [DYNAMIC_EXECUTION]: The skill provides commands to execute arbitrary GDScript code directly via the CLI (
cli-anything-godot script inline) or by running existing script files (cli-anything-godot script run). This allows for dynamic execution of logic within the Godot environment. - [COMMAND_EXECUTION]: The skill facilitates shell command execution to interact with the Godot engine, including project creation, resource re-importing, and build exports.
- [INDIRECT_PROMPT_INJECTION]: The skill reads external data from Godot projects (scenes, scripts, and resources) which could contain malicious instructions designed to influence the agent's actions.
- Ingestion points: Reading scene structures (
scene read), listing project scripts (project scripts), and project scenes (project scenes). - Boundary markers: None identified in the prompt templates or command descriptions.
- Capability inventory: File creation (
project create,scene create), scene modification (scene add-node), and arbitrary code execution (script run,script inline). - Sanitization: No evidence of sanitization or validation of the content read from Godot files before it is processed by the agent.
Audit Metadata