getting-started-with-amplitude
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references supplemental resources and skills hosted on the vendor's official GitHub repository (amplitude/builder-skills). These are part of the intended product ecosystem for existing customers.
- [COMMAND_EXECUTION]: Instructs the agent to invoke the
get_amplitude_contexttool to verify user permissions and retrieve project metadata from the Amplitude platform. - [INDIRECT_PROMPT_INJECTION]: The skill establishes a data ingestion surface by connecting the agent to Amplitude's Product MCP servers for data analysis.
- Ingestion points: Project data, charts, and dashboards retrieved from
https://mcp.amplitude.com/mcporhttps://mcp.eu.amplitude.com/mcp. - Boundary markers: Refers to a session-wide
SAFETY.mdfile for expectation setting. - Capability inventory: Capabilities include project context retrieval, routing to chart/dashboard creation tools, and experiment monitoring.
- Sanitization: Sanitization of external data is not explicitly defined in this onboarding instruction file.
Audit Metadata