doc-bdd-autopilot_skill

This skill generates ADR-ready BDD scenarios from EARS requirements, validates readiness scores, and categorizes outcomes to accelerate documentation.

Python

9

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill vladm3105/aidoc-flow-framework --skill doc-bdd-autopilot

SKILL.md53.2 KB

Overview

This skill automates generation and review of BDD (Gherkin) scenarios from EARS requirement documents and enforces ADR-Ready scoring. It orchestrates readiness checks, sectioned feature creation, scenario categorization, and iterative validation with auto-fix actions when possible. The pipeline produces traceable, tagged feature files and a validation-ready ADR score.

How this skill works

The autopilot detects input type (EARS or BDD) and either generates a BDD suite from EARS sources or runs a review on an existing BDD suite. It runs a BDD-Ready validation on EARS, plans a sectioned feature layout, generates success/error/edge/data-driven/quality scenarios using Gherkin patterns, and applies real-time quality feedback. After generation it validates ADR-Ready scoring and applies fixes until thresholds are met.

When to use it

You have completed EARS documents and need automated Gherkin scenario generation
You want consistent scenario categorization: success, error, edge, data-driven, quality attributes
You need BDD-Ready and ADR-Ready score enforcement (configurable minimums)
You want traceable, tag-rich feature files with BRD/PRD/EARS links
You need bulk generation or review across multiple EARS/BDD inputs

Best practices

Run EARS validation first and ensure BDD-Ready score ≥ 90% before generation
Keep EARS statements atomic and quantifiable to improve testability and reduce auto-fix edits
Use the autopilot for bulk or standard conversion; for heavy manual customization, edit generated files directly
Provide sectioned EARS documents or clear statement counts so the planner can create sensible feature sections
Validate threshold references in PRD files so @threshold tags resolve correctly during generation

Example use cases

Generate BDD suites for multiple EARS inputs and produce ADR-ready feature files for engineering handoff
Review an existing BDD-NN suite to produce a quality and ADR-Ready validation report
Create data-driven Gherkin scenarios from parameterized EARS statements with Examples tables and threshold tags
Auto-fix common EARS issues (non-atomic statements, missing constraints) and re-run generation to meet readiness score
Add cumulative traceability tags (@brd, @prd, @ears) across all generated features for downstream traceability

FAQ

Input is identified by document type codes like EARS-NN or BDD-NN. EARS inputs trigger generation; BDD inputs trigger review.

What readiness scores are required?

Default minimums are 90% for BDD-Ready and ADR-Ready. These thresholds are configurable but recommended to ensure quality.

Can the autopilot fix issues automatically?

Yes. It can apply defined auto-fix actions (reformatting, splitting statements, adding placeholders) and re-validate until scores meet thresholds.