postmark-automation
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: Indirect prompt injection surface identified. Ingestion points: The skill processes untrusted data through TemplateModel variables in POSTMARK_SEND_BATCH_WITH_TEMPLATES and template content in POSTMARK_EDIT_TEMPLATE as described in SKILL.md. Boundary markers: Absent; there are no delimiters or instructions to the agent to treat these variables as untrusted or to ignore potentially malicious embedded instructions. Capability inventory: The skill includes significant capabilities such as sending batch emails and modifying server settings (e.g., POSTMARK_SEND_BATCH_WITH_TEMPLATES, POSTMARK_EDIT_SERVER). Sanitization: Absent; no sanitization, validation, or escaping of template inputs is mentioned.
- [EXTERNAL_DOWNLOADS]: The skill references an external MCP server at https://rube.app/mcp which provides the tool definitions and logic required for the Postmark automation workflows.
Audit Metadata