aming-claw-hn-demo-before-work

To collect structured evidence before AI-assisted code reviews, you must follow a fixed pre-work sequence: read MCP resources, verify project context, collect graph-first evidence, and confirm backlog and fence details.

Graph-first discovery in AI demo governance is the process of starting from a project's structure, ownership, and neighbors to anchor planning, ensuring safe and deterministic AI demonstrations by preventing structural misinterpretations.

Backlog contracts and target file fences are required for AI-assisted reviews to prevent duplicate work and misinterpretations. They enforce governance-aware workflows by capturing exact backlog evidence, fence details, and acceptance criteria before implementation begins.

You enforce a deterministic pre-work sequence for graph-first AI demos by applying governance guardrails that require reading MCP resources, verifying project context, collecting evidence, and confirming backlog and fence details before any implementation starts.

Yes, you can apply acceptance criteria governance for AI-assisted code reviews without external dependencies. The skill operates standalone to enforce pre-work sequences, collect graph-first evidence, and ensure deterministic results without requiring additional components.

The limitation of using pre-work guardrails for AI demos is the strict enforcement overhead. You must complete the full evidence collection sequence—reading MCP resources, verifying context, and confirming fences—before any implementation, which may slow down rapid prototyping.