typo3-idea-extension-blog
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches external content to drive the extension building and blogging process. 1. Ingestion points: Phase 1 (Fetch the full article, announcement, or product page). 2. Boundary markers: Not specified in instructions to prevent the model from obeying instructions found in fetched content. 3. Capability inventory: File-write (Phase 4), Git push (Phase 5). 4. Sanitization: Not specified for external content processing.
- [COMMAND_EXECUTION]: The skill utilizes Git commands to manage and push implementation code. This is a core feature and includes explicit instructions for credential cleanup.
- [EXTERNAL_DOWNLOADS]: The skill fetches external web content for research and evaluation purposes. This behavior is necessary for the intended functionality and targets user-provided or independent research URLs.
Audit Metadata