signals-scout-experiments
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data that can be influenced by external users, creating a potential surface for indirect prompt injection. \n
- Ingestion points: Data from
experiment-get(titles and descriptions) andadvanced-activity-logs-list(user activity messages) are incorporated into the agent's context. \n - Boundary markers: The instructions do not specify the use of delimiters or clear boundaries when interpolating these fields into reports or analysis. \n
- Capability inventory: The agent has access to
execute-sqlfor querying the database andemit_report/edit_reportfor broadcasting findings, which could be abused if the agent is manipulated by injected instructions. \n - Sanitization: There are no instructions for sanitizing or escaping the content retrieved from experiment metadata or activity logs before it is used.
Audit Metadata