amplitude-automation
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires connecting to the Rube MCP server endpoint at https://rube.app/mcp and references the Composio toolkit documentation at https://composio.dev. These are established service providers for the skill's primary functionality.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external Amplitude sources which could present a surface for indirect prompt injection. • Ingestion points: Data is ingested from the Amplitude API via tools such as AMPLITUDE_GET_USER_ACTIVITY, AMPLITUDE_FIND_USER, and AMPLITUDE_LIST_COHORTS as described in SKILL.md. • Boundary markers: The skill does not specify boundary markers or instructions to ignore embedded commands within the ingested data. • Capability inventory: The skill possesses write capabilities including AMPLITUDE_SEND_EVENTS and AMPLITUDE_UPDATE_COHORT_MEMBERSHIP in SKILL.md. • Sanitization: There is no evidence of specific sanitization or filtering logic applied to the data retrieved from external sources.
Audit Metadata