seo-article-pipeline
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill demonstrates secure orchestration patterns, relying on standard platform tools for file management and task delegation. Configuration data consists of organizational resource identifiers (Google Drive IDs and a workspace email) rather than sensitive credentials or secrets.\n- [PROMPT_INJECTION]: The skill ingests untrusted brand and article material packets and passes them as input to downstream sub-agents. Although this workflow presents a surface for indirect prompt injection, the risk is effectively mitigated by the skill's design, which requires explicit user confirmation (HITL) for all handoffs, preventing automated execution of potentially malicious instructions embedded in the input data.
Audit Metadata