skills/romiluz13/ddd/ddd-ground/Gen Agent Trust Hub

ddd-ground

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill exposes the agent to untrusted external data which could serve as a vector for indirect prompt injection.\n
  • Ingestion points: The instructions direct the agent to "Consult new APIs and research failed assumptions," necessitating the ingestion of external documentation and research results into the agent's context.\n
  • Boundary markers: There are no instructions or structural markers defined to help the agent distinguish between passive data in external sources and potential instructions embedded within those sources.\n
  • Capability inventory: The skill enables "authorized implementation," which involves the generation and execution of code snippets and file modifications based on the grounded plan.\n
  • Sanitization: The skill lacks requirements for sanitizing or validating the ingested external data before the agent uses it to drive implementation or decision-making.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 11:20 PM
Security Audit — agent-trust-hub — ddd-ground