typo3-powermail

Pass

Audited by Gen Agent Trust Hub on Oct 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documents a system designed to process user-supplied form data, creating an attack surface for indirect injection (SKILL.md, SKILL-EXAMPLES.md). Ingestion points: Data enters the agent context through various form field types such as text inputs, textareas, and file uploads. Boundary markers: The skill details the use of Powermail's built-in validation framework and SpamShield methods (e.g., HoneyPod, LinkMethod) to define input constraints. Capability inventory: Submissions can trigger database persistence (SaveToAnyTableFinisher), file system operations (UploadValidator), and external network requests via custom finishers (04-custom-finishers.md). Sanitization: Input is processed through established TYPO3 validators and PSR-14 event listeners.
  • [COMMAND_EXECUTION]: The skill includes numerous DDEV CLI commands and raw SQL scripts (SKILL-EXAMPLES.md) for environment configuration and database population. These instructions are explicitly scoped for local development use only and include warnings against production execution.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to install a vendor-maintained fork from a GitHub VCS repository (https://github.com/dirnbauer/powermail) to facilitate compatibility with TYPO3 v14.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 2, 2026, 09:28 AM