test-master
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill's frontmatter identifies the author as 'Jeffallan' (via GitHub profile link), which contradicts the assigned author context of 'emmraan'. This discrepancy represents deceptive metadata that could lead to misjudgment of the skill's provenance.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and analysis of user-provided source code to generate and execute tests, establishing a vulnerability surface where malicious instructions in the code could influence agent behavior.
- Ingestion points: Source code and project files analyzed for the creation of unit, integration, and E2E tests as specified in
SKILL.mdandreferences/qa-methodology.md. - Boundary markers: Absent; there are no instructions provided to the agent to use delimiters or to disregard natural language instructions found within the code being tested.
- Capability inventory: The skill instructions involve the execution of shell-based test runners (Jest, Playwright, k6) and file-system write operations for generating test architectures across all reference files.
- Sanitization: The skill lacks validation or sanitization protocols for handling external code content before it is interpolated into test generation prompts.
Audit Metadata