aibdd-discovery

Execute AIBDD discovery SOP phases to produce boundary-aware artifacts from a plan spec.

77|15|Updated May 12, 2026
One-click install
npx skills add https://github.com/Waterball-Software-Academy/aixbdd --skill aibdd-discovery-waterball-software-academy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aibdd-discovery
Source: https://github.com/Waterball-Software-Academy/aixbdd/tree/main/.agents/skills/aibdd-discovery
Command: npx skills add https://github.com/Waterball-Software-Academy/aixbdd --skill aibdd-discovery-waterball-software-academy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

AIBDD discovery prevents AI agents from guessing specs by converting raw product intent into structured, boundary-aware artifacts and explicit next-phase guidance.

Core Features & Use Cases

  • SOP-driven discovery execution: Runs the required discovery phases in a strict order to produce discovery sourcing and update the plan spec as truth.
  • CWD-anchored artifact safety contract: Enforces that all created/modified artifacts remain within the current working directory scope.
  • Atomic rule extraction and rule hygiene: Enumerates atomic rules per feature file and identifies what must be fixed or clarified, delegating clarification to /clarify-loop when needed.
  • Impact-scope maintenance: Maintains a machine-readable impact matrix so downstream planning knows exactly what boundary truths were affected and how.

Quick Start

Run the aibdd-discovery skill in your project working directory after completing /aibdd-kickoff so that arguments.yml exists, then review the produced plan reports and proceed to aibdd-plan.

Frequently Asked Questions about aibdd-discovery

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert raw product intent into structured specification boundaries?

Boundary-aware discovery artifacts are generated by executing discovery SOP phases sequentially from a plan spec. They enumerate atomic rules per feature file, detect specification gaps, and delegate clarification to maintain an impact matrix for downstream planning.

How do I extract atomic rules and acceptance criteria from a plan spec?

Extracting atomic rules and acceptance criteria from a plan spec requires running an SOP-driven discovery phase. This process enumerates rules per feature file, identifies gaps requiring fixes, and updates the plan spec as the single source of truth.

Do I need to run aibdd-kickoff before executing the discovery workflow?

Yes, you must run aibdd-kickoff before executing the discovery workflow to generate the required arguments.yml file. The discovery process relies on resolving required arguments from this configuration to execute its SOP phases correctly.

Can I generate discovery artifacts outside the current working directory?

No, you cannot generate discovery artifacts outside the current working directory due to a strict CWD-anchored artifact safety contract. All created or modified artifacts, including discovery-sourcing.md and plan reports, must remain within the current working directory scope.

What happens when the discovery process finds insufficient evidence for acceptance criteria?

When the discovery process finds insufficient evidence for acceptance criteria, it delegates clarification to the clarify-loop. This ensures that specification gaps and ambiguous atomic rules are explicitly resolved rather than guessed by the AI agent.

How does the discovery phase track changes for downstream planning?

The discovery phase tracks changes for downstream planning by maintaining a machine-readable impact matrix via the manage_impact_matrix wrapper. This matrix records exactly which boundary truths were affected and how they influence future phases.