agent-explorer-role

To define scope and gather evidence for AI-driven validation tasks, establish clear boundaries, assumptions, and unknowns before execution. Collect comprehensive checks across happy paths, error paths, and boundaries to ensure all outputs are supported by anchors or counter-checks.

An output contract for explorer roles in AI-driven tasks is a structured reporting format. It mandates that findings include a final verdict, verified information, and a summary of facts and risks, ensuring all conclusions are verifiable.

Structure validation reports by adhering to a defined output contract that separates the verdict, verified information, and a summary of facts and risks. This ensures structured, verifiable reporting supported by evidence anchors and counter-checks.

Enforce mandatory explorer role guidelines when an explorer needs to validate a new feature. This ensures they meticulously define testing boundaries, gather necessary evidence across paths, and report findings in a structured, verifiable manner.

Before executing AI-driven tasks, define what is explicitly in and out of scope, along with all assumptions and unknowns. This scope definition phase is mandatory to guide subsequent comprehensive evidence gathering and validation.

Evidence gathering for AI tasks requires checking error paths and boundaries. It mandates comprehensive collection across happy paths, error paths, and boundaries, with each output supported by anchors or counter-checks for verifiable validation.