The Agent Skills Directory

[SAFE]: The skill contains no executable code, shell commands, or network operations. All instructions are focused on providing architectural guidance and generating documentation.
[INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data to recommend testing strategies.
Ingestion points: Reads project architecture and UI/API characteristics provided in the agent's context.
Boundary markers: None explicitly defined in the instructions, though the output is confined to a markdown template.
Capability inventory: The skill only generates a markdown file (TEST_PLAN.md) and provides text-based recommendations. It lacks the ability to execute shell commands, write to arbitrary file paths, or make network calls.
Sanitization: Not applicable as the output is inert text for human or agent review.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns or package installations were detected. The skill mentions industry-standard testing tools (e.g., Vitest, Playwright, Pact) as recommendations for the user to implement, rather than attempting to install or run them itself.
[DATA_EXFILTRATION]: No data exfiltration patterns detected. The skill includes a dedicated security section warning against routing production data through third-party CI environments, demonstrating a security-conscious design.

testing-strategy