html-mailer-builder
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted user data to populate email templates.
- Ingestion points: User-provided requirements such as company name, content details, and URLs are written to
config.json(as described inSKILL.md). - Boundary markers: The skill does not explicitly instruct the agent to use delimiters or ignore instructions embedded within the user-provided content during the requirement gathering phase.
- Capability inventory: The skill executes a local Python script (
scripts/generate_email_template.py) to process data and write files. - Sanitization: The
scripts/generate_email_template.pyscript mitigates risks by usinghtml.escape()for text fields before inserting them into the HTML structure. - [COMMAND_EXECUTION]: The workflow requires the agent to execute a local Python script to perform its primary function.
- Evidence:
python scripts/generate_email_template.py --input config.json --output email.htmlas documented in the Workflow section ofSKILL.md. - Observation: The execution is scoped to a local script provided within the skill and is necessary for the tool's intended utility.
Audit Metadata