[WIP] Update compile command to show failed workflow IDs

[Copilot]

Plan: Agent-Friendly Compile Error Summary

• Understand the current implementation
• Identify the problem: Summary doesn't clearly list failed workflow IDs before showing detailed errors
• Update printCompilationSummary to show failed workflow list immediately after summary line
• Add tests to validate the new output format
• Test manually with multiple failing workflows
• Run make agent-finish to validate changes

Summary

Updated the compile command to provide an agent-friendly summary of failed workflows. When compilation errors occur, the output now shows:

1. Overall summary line: ✗ Compiled X workflow(s): Y error(s), Z warning(s)
2. Clear list of failed workflow IDs:

   ✗ Failed workflows:
     ✗ workflow1.md
     ✗ workflow2.md
   

3. Detailed error messages for each workflow

This makes it much easier for agents (and humans) to quickly identify which workflows failed before diving into the detailed error messages.

Changes Made

• Updated printCompilationSummary() in pkg/cli/compile_helpers.go to display failed workflow list immediately after the summary line
• Added comprehensive unit tests in pkg/cli/compile_summary_output_test.go validating the new output format
• Maintains backward compatibility with the existing FailedWorkflows field
• All tests passing, code formatted and linted

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[Copilot]

Pull request overview

Updates the compile command’s stderr summary so failed workflow IDs are listed immediately after the overall compilation summary line, before detailed per-workflow errors.

Changes:

• Print a “Failed workflows” section (with workflow IDs) right after the error summary line when compilation fails.
• Keep backward compatibility by still supporting FailedWorkflows when FailureDetails is absent.
• Add unit tests to validate the updated output.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

File	Description
pkg/cli/compile_helpers.go	Prints a dedicated “Failed workflows” list (and spacing) before detailed error messages.
pkg/cli/compile_summary_output_test.go	Adds tests capturing stderr to validate summary + failed workflow list output.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

This test mutates the global os.Stderr, which can race with other tests (including tests in other packages) if they run in parallel. Prefer refactoring printCompilationSummary to write to an injected io.Writer (e.g., parameter or package-level writer) so tests can pass a bytes.Buffer without touching global process state; if keeping this approach, at minimum restore os.Stderr via t.Cleanup/defer, check the os.Pipe()/ReadFrom errors, and close r to avoid leaking FDs.

Suggested change

	r, w, _ := os.Pipe()
	os.Stderr = w
	// Call the function
	printCompilationSummary(tt.stats)
	// Restore stderr and capture output
	w.Close()
	os.Stderr = oldStderr
	var buf bytes.Buffer
	buf.ReadFrom(r)
	r, w, err := os.Pipe()
	if err != nil {
	t.Fatalf("failed to create pipe: %v", err)
	}
	os.Stderr = w
	t.Cleanup(func() {
	os.Stderr = oldStderr
	})
	// Call the function
	printCompilationSummary(tt.stats)
	// Restore stderr and capture output
	os.Stderr = oldStderr
	if err := w.Close(); err != nil {
	t.Fatalf("failed to close write end of pipe: %v", err)
	}
	var buf bytes.Buffer
	if _, err := buf.ReadFrom(r); err != nil {
	t.Fatalf("failed to read from pipe: %v", err)
	}
	if err := r.Close(); err != nil {
	t.Fatalf("failed to close read end of pipe: %v", err)
	}

[Copilot]

The new behavior is specifically about ordering (failed workflow list appearing immediately after the summary line and before detailed errors), but these assertions only check presence. Add ordering assertions (e.g., using strings.Index to ensure Compiled ... occurs before Failed workflows: and that occurs before the first detailed error message) so the test will fail if the output sections are reordered.

[Copilot]

In the FailedWorkflows fallback branch you print workflow as-is, but in the FailureDetails branch you print filepath.Base(failure.Path). For consistent output (and to better match the stated goal of listing workflow IDs), consider applying filepath.Base(workflow) here as well.

Suggested change

	fmt.Fprintf(os.Stderr, " ✗ %s\n", workflow)
	fmt.Fprintf(os.Stderr, " ✗ %s\n", filepath.Base(workflow))