amplitude
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external Amplitude analytics sources, which could potentially contain embedded instructions in event names, feedback surveys, or user activity logs.
- Ingestion points: External data enters the agent context through the use of Amplitude MCP tools for search and retrieval (SKILL.md).
- Boundary markers: The instructions do not define specific boundary markers or delimiters for encapsulating untrusted external data within prompts.
- Capability inventory: The skill is restricted to read-only tool invocations and contains explicit prohibitions against using shell commands or modifying Amplitude resources (SKILL.md).
- Sanitization: The skill provides instructions to return only the fields necessary for the request and to distinguish processed themes from raw customer comments (SKILL.md).
- [SAFE]: The skill consists entirely of markdown documentation and YAML frontmatter. It does not include executable code or scripts and enforces strict adherence to read-only analytics workflows using legitimate vendor resources.
Audit Metadata