sdd-design

Pass

Audited by Gen Agent Trust Hub on Mar 28, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it ingests and processes technical artifacts that may be controlled by external parties.\n
  • Ingestion points: Technical proposals and specifications are retrieved via the mem_get_observation tool from an internal store or read directly from the filesystem (e.g., proposal.md).\n
  • Boundary markers: The instructions do not provide explicit boundary markers or directives for the agent to treat the content of these external artifacts as untrusted data.\n
  • Capability inventory: The skill possesses capabilities to write to the local filesystem (creating design.md in the openspec/ directory) and to persist artifacts in the project's artifact store via the mem_save tool.\n
  • Sanitization: No sanitization or validation logic is defined to inspect the ingested specification content before it is incorporated into the design process.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 28, 2026, 02:23 PM