define a speckit-based bdd agent

[adrian-herscu]

Is your feature request related to a problem? Please describe.

Test and automation developers (the user) are required to come up with various tests and have them implemented as readable scenarios which can be executed by a computer. This requires a lot of time and skill to implement correctly and maintain.

Describe the solution you'd like

I want to support BDD scenario generation using Copilot in VSCode (the AI) via speckit-like prompts and agents.
The user opens the Chat in VSCode and starts a BDD workflow comprising of following prompts:

1. /bdd.specify <user's goal>

a. the AI shall:
a.1. interact with the system under test (SUT) using appropriate MCP server(s) and/or CLI commands
a.2. validate that user's goal is achievable within a reasonable time-window, and if not, shall asks the user for clarifications, perhaps also asking the user to install additional software, check connectivity, authentication, etc.

b. upon successful validation, the AI shall present a high-level BDD scenario in Gherkin format for the user to approve

c. upon user's approval, the AI shall generate a skeleton test method adding the approved BDD scenario as method-level documentation

d. the skeleton test method should be integrated into existing test suite and reuse existing test infrastructure as appropriate

2. /bdd.design

the AI shall implement the skeleton test method, preserving the abstraction-level in the approved BDD scenario

3. /bdd.implement

the AI shall implement the steps of the test method reusing existing steps as appropriate

NOTE: interacting with a SUT may require authentication; the credentials shall not be considered secret since the SUT itself is in a test environment