sdd-tasks

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from design and specification documents to generate implementation tasks, creating an attack surface for indirect prompt injection.
  • Ingestion points: The skill reads content from sdd/{change-name}/proposal, sdd/{change-name}/spec, and sdd/{change-name}/design (Execution and Persistence Contract).
  • Boundary markers: The instructions do not define explicit boundary markers or specific guidance to ignore instructions that might be embedded within the ingested design documents.
  • Capability inventory: The skill has file-writing capabilities, specifically for creating the tasks.md file (Step 3). It does not demonstrate network or subprocess execution capabilities.
  • Sanitization: There is no explicit sanitization, filtering, or validation defined for the content extracted from the input design artifacts before it is interpolated into the task breakdown.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 03:17 PM
Security Audit — agent-trust-hub — sdd-tasks