content-strategist
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from a research corpus, which creates a surface for indirect prompt injection.
- Ingestion points: The core workflow in
SKILL.mdbegins by reading a complete 'knowledge map' from theresearch-intakeskill. - Boundary markers: The instructions do not define any delimiters or specific 'ignore' rules for the ingested research data, increasing the risk that instructions embedded in the data could influence agent behavior.
- Capability inventory: The skill generates markdown plans and tracking tables and interacts with users via
AskUserQuestion. It does not have access to sensitive files, network operations, or shell tools. - Sanitization: No validation or sanitization processes are defined for the external research data before it is used to build content plans.
Audit Metadata