typespec-create-agent

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a structured template and guidelines for creating Microsoft 365 Copilot agents using official TypeSpec decorators.
  • [SAFE]: The dependencies referenced in the code snippets (such as @microsoft/typespec-m365-copilot, @typespec/http, and @typespec/openapi3) are well-known, official libraries from Microsoft.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides a template that interpolates user-provided instructions into an agent definition, creating a potential attack surface for indirect prompt injection if the user provides malicious content during the generation process.
  • Ingestion points: User input defining the agent's purpose, role, and behavioral guidelines via the 'Examples' workflow in SKILL.md.
  • Boundary markers: The template uses triple-quoted strings for the @instructions decorator, which provide basic structural separation for the injected content.
  • Capability inventory: The generated agent can be configured with powerful capabilities including Email, OneDriveAndSharePoint, Meetings, and CodeInterpreter.
  • Sanitization: No explicit sanitization or validation of the user-provided text is performed or described in the skill's instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 02:17 AM
Security Audit — agent-trust-hub — typespec-create-agent