test-scaffolding
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted code snippets from users to generate test scaffolding, exposing the agent to potentially malicious instructions embedded in comments or logic.
- Ingestion points: Untrusted source code provided by users when requesting tests or showing new functionality as described in the activation triggers in SKILL.md.
- Boundary markers: The skill does not define or use explicit delimiters (e.g., specific XML tags or triple backticks with ignore-instructions) to separate user data from the agent's logic.
- Capability inventory: The skill facilitates the generation of executable test code for multiple platforms (Drupal, WordPress, React).
- Sanitization: The instructions do not include steps to sanitize, filter, or escape input code to prevent the agent from following instructions embedded within the user's data.
Audit Metadata