testcase-writer-plus
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external requirement and analysis documents to generate test cases. This creates an inherent surface area for indirect prompt injection if a source document contains malicious instructions intended to manipulate the agent's behavior. However, the risk is low as the skill lacks network access or privilege escalation capabilities. * Ingestion points: The skill uses scripts/common_parser.py and scripts/xlsx_reader.py to extract text from .docx, .xlsx, .html, and .json files. * Boundary markers: While the prompt in prompts/testcase-writer-plus.md is highly structured, it does not employ specific delimiters or 'ignore' instructions to encapsulate source content from the documents. * Capability inventory: The skill is limited to reading local source files and writing formatted output to the local filesystem. * Sanitization: Extracted text is used directly without semantic filtering for malicious directives.
- [SAFE]: The Python scripts provided are well-written and use only standard library modules. No external dependencies or remote code executions are present. The skill's architecture explicitly separates the deterministic preparation phase from the semantic analysis phase, reducing the risk of automated malicious execution.
Audit Metadata