Using SPECTRA Execution Agent with Copilot Chat

Overview

The SPECTRA Execution Agent enables interactive test execution through GitHub Copilot Chat in VS Code.

Setup

1. Initialize SPECTRA in your repository:

   spectra init
   

2. This creates the agent file at .github/agents/spectra-execution.agent.md

3. Open VS Code with the repository

4. Ensure you have GitHub Copilot extension installed

Invocation

Open Copilot Chat and invoke the agent:

@spectra-execution run tests

Or with a specific suite:

@spectra-execution run checkout tests

Workflow

1. Suite Selection: The agent lists available test suites
2. Filter Options: Choose priority, tags, or component filters
3. Test Case Presentation: Each test case is presented one at a time with:
   • Title and priority
   • Preconditions
   • Numbered steps
   • Expected result
   • Test data (if any)
4. Result Collection: Say “passed”, “failed”, “blocked”, or “skip”
5. Progress Tracking: See live progress after each test
6. Summary: Final pass/fail counts when the run completes

Example Session

You: @spectra-execution run auth tests

Agent: I found 3 test suites:
- auth (18 test cases)
- checkout (42 test cases)
- orders (7 test cases)

Which suite would you like to run? Any filters?

You: auth, high priority only

Agent: Starting run with 5 high-priority test cases...

## TC-101: Login with valid credentials
**Priority**: high | **Component**: auth

### Steps
1. Navigate to the login page
2. Enter valid email "test@example.com"
3. Enter valid password "SecurePass123"
4. Click Sign In

### Expected Result
User is redirected to dashboard

---
**Progress**: Test 1/5
What is the result?

You: passed

Agent: ✓ TC-101 passed. Progress: 1/5 — 1 passed
[presents next test...]

Tips

• Use natural language for results (“it worked” = pass, “bug found” = fail)
• The agent will ask for failure comments before moving on
• Test state is preserved if connection is lost