The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional Markdown content designed to improve the agent's reasoning about software testing. It contains no code, binary files, or automated commands.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to analyze untrusted input such as code files, specifications, and user tickets. However, this is inherent to its intended purpose, and the risk is minimal because the skill's output is limited to natural language test plans rather than code execution.
Ingestion points: Phase 1 of SKILL.md instructs the agent to read implementation artifacts and business rule sources provided by the user.
Boundary markers: There are no specific delimiters defined to wrap or isolate the untrusted data within the context.
Capability inventory: The skill relies on standard file system reading and searching capabilities.
Sanitization: No explicit sanitization or filtering steps are required by the instructions beyond standard AI processing.

test-design