azure-typespec-author

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through its core operations of reading user input and existing file content.\n
  • Ingestion points: Project files (e.g., .tsp, tspconfig.yaml) and user requests are ingested during the intake and planning steps defined in SKILL.md and references/intake-arm.md.\n
  • Boundary markers: The skill instructions do not explicitly require the use of boundary markers or delimiters to isolate untrusted data from the system prompt.\n
  • Capability inventory: The skill allows modification of files in the local repository and the execution of validation tools via the MCP interface.\n
  • Sanitization: The workflow relies on the correctness of a generated authoring plan without explicit safety-specific sanitization of the underlying data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 08:14 AM
Security Audit — agent-trust-hub — azure-typespec-author