manual-test-script-author
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input data (user stories, PRD sections, bug reports, and exploratory debriefs) to generate manual test documentation. The instructions do not specify the use of boundary markers or sanitization steps to isolate this data, which could potentially allow malicious instructions within the source material to influence the agent's output.
- Ingestion points: User story/PRD sections, bug reports, exploratory testing debriefs, and acceptance criteria line items (SKILL.md, Step 1).
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore embedded instructions in the source data.
- Capability inventory: The skill performs markdown text generation for various test formats (step-tables, Gherkin scenarios, UAT scripts, and checklists).
- Sanitization: No procedures for escaping or filtering external content are mentioned in the skill files.
Audit Metadata