email-render-builder
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and process externally provided HTML, email templates, and creative copy, which presents a surface for indirect prompt injection. Malicious instructions could be hidden in the processed data to influence the agent's behavior.
- Ingestion points: Raw HTML snippets, exported email templates, and creative creative content provided as user input in
SKILL.md. - Boundary markers: The skill contains a strong explicit warning to treat pasted HTML as untrusted input and to disregard any instructions embedded within it, referencing a standard security policy.
- Capability inventory: The skill can execute a local Python connector script (
resend.py) and write detailed render-QA reports to memory files. - Sanitization: The skill relies on the agent's adherence to natural language instructions to ignore embedded commands; no automated sanitization is specified for the input text.
- [COMMAND_EXECUTION]: The skill facilitates the execution of a local Python script to perform live render tests of the built HTML.
- Evidence: The instructions in
SKILL.mddescribe usingpython3 "${CLAUDE_PLUGIN_ROOT}/scripts/connectors/resend.py"to deliver test emails. This is a controlled execution of an internal plugin utility for verifying client rendering.
Audit Metadata