agents-sdk
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- External Resource Management: The skill downloads configuration and deployment tools from official repositories, specifically the OpenAI cookbook on GitHub. This is used to facilitate local deployments using standard infrastructure.
- Development Operations: The skill automates project setup and testing using local tools like git, make, and uv. These operations are standard for managing the application lifecycle and dependency installation.
- Instruction Ingestion Considerations: The skill processes content from external sources such as repository files and interaction threads to generate code. Ingestion points: Target repository files (README.md, entrypoints) and Codex thread data (SKILL.md). Boundary markers: None identified to delimit external content from instructions. Capability inventory: Execution of shell commands via make, uv, git, and curl (SKILL.md). Sanitization: No explicit filtering or sanitization of ingested content was observed. This represents a standard interface for code-generation skills operating on user-provided context.
- Application Execution Workflow: A core functionality involves generating and running scripts (e.g., for diagram generation or app entrypoints). This dynamic execution is essential for the skill's primary purpose of creating runnable prototypes.
Audit Metadata