skills/google/adk-docs/adk-scaffold/Gen Agent Trust Hub

adk-scaffold

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • External Script Execution: The documentation provides a troubleshooting step that downloads and executes an installation script for the uv tool from its official domain. This is a common installation pattern for this developer utility.
  • Remote Tool Execution: The skill utilizes uvx to fetch and execute the agent-starter-pack tool directly from a package registry. This ensures the agent uses the most current version of the scaffolding logic provided by the tool's author.
  • Command Execution: The skill's primary function involves running CLI commands to manage project files and infrastructure, requiring the ability to execute subprocesses.
  • Indirect Prompt Injection Surface: The skill processes user-defined project requirements and includes them in generated commands and specifications. (1) Ingestion points: User input for project requirements gathered in Step 1 of SKILL.md. (2) Boundary markers: None identified in the provided templates. (3) Capability inventory: Execution of uvx, curl, and pip as described in SKILL.md. (4) Sanitization: No specific validation or escaping of user-provided project metadata is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 09:06 PM