compare-user-journeys
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of untrusted data from session events and user feedback.
- Ingestion points: Untrusted data enters the agent context via
Amplitude:get_session_replay_events(Step 5) andAmplitude:get_feedback_insights(Step 6) as described inSKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to isolate the external content from the agent's core logic.
- Capability inventory: The skill's primary capabilities include querying metrics, summarizing interactions, and generating journey maps and hypotheses. It lacks high-privilege commands like system execution or file-writing, limiting the impact of an injection.
- Sanitization: There is no evidence of content sanitization or filtering applied to the interaction timelines or feedback themes before they are analyzed by the model.
Audit Metadata