skills/microsoft/vscode/component/Gen Agent Trust Hub

component

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection Surface]: The skill is designed to ingest context from the user and repository data to generate architectural overviews. This ingestion process is a standard surface for indirect prompt injection, where data contained within the repository could attempt to influence the agent's behavior.
  • Ingestion points: The skill reads user-provided context and uses tools to gather information from the codebase.
  • Boundary markers: The instructions do not define specific boundary markers or "ignore" instructions for the generated markdown output to separate gathered content from the agent's core instructions.
  • Capability inventory: The skill utilizes file system capabilities to read repository data and create new documentation files in the .components/ directory.
  • Sanitization: The instructions do not explicitly require the validation or sanitization of the content gathered from the repository before writing it to the new component overview file.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 07:52 PM
Security Audit — agent-trust-hub — component