[Refactoring] Split src/commands/main-action.test.ts into focused test suites

[github-actions]

Refactoring Opportunity

Summary

• File: src/commands/main-action.test.ts
• Current size: 1,173 lines (22 describe blocks, 54 it cases)
• Responsibilities identified: at least 4 distinct concerns bundled into one spec file

Evidence

Top-level describe blocks span unrelated concerns:

80:describe('createMainAction', () => {
114:  describe('when args is empty', () => {
151:  describe('when --reflect is used with a command', () => {
162:  describe('when single arg is provided', () => {
173:    describe('work directory selection', () => {
256:  describe('when multiple args are provided', () => {
273:  describe('happy path', () => {
424:  describe('when runMainWorkflow returns non-zero exit code', () => {
432:    describe('sbx runtime wiring', () => {
481:  describe('when runMainWorkflow throws', () => {
590:    describe('when external runtime resolution fails', () => {
609:    describe('when external runtime preflight fails', () => {
678:  describe('performCleanup with keepContainers=true', () => {
727:  describe('external runtime cleanup failures', () => {
759:  describe('external runtime diagnostics', () => {
798:  describe('performCleanup with containers started', () => {
828:  describe('performCleanup signal parameter', () => {
845:  describe('onContainersStarted and onHostIptailesSetup callbacks', () => {
866:  describe('fatal error cleanup after containers started', () => {
900:  describe('redaction of sensitive config fields', () => {

followed by a describe('resolved config artifact', ...) block (audit-dir writing, secret redaction in the on-disk artifact) that runs to the end of the file.

These map onto at least four logically separate concerns: (1) CLI arg parsing/dispatch (empty args, --reflect, single/multi args, work-dir selection), (2) container lifecycle / external runtime wiring (sbx runtime, external runtime resolution/preflight failures), (3) cleanup semantics (performCleanup variants, signal handling, callbacks, fatal-error cleanup), and (4) secret redaction / audit artifact writing (both the "redaction of sensitive config fields" and "resolved config artifact" blocks).

Proposed Split

• src/commands/main-action.test.ts — CLI dispatch and arg-parsing behavior (~300 lines)
• src/commands/main-action-lifecycle.test.ts — container/external-runtime start, sbx wiring, resolution/preflight failure paths (~350 lines)
• src/commands/main-action-cleanup.test.ts — performCleanup variants, signal handling, callbacks, fatal-error cleanup (~250 lines)
• src/commands/main-action-audit.test.ts — secret redaction and awf-resolved-config.json artifact writing (~270 lines)

Shared fixtures (e.g. MAIN_ACTION_STUB_CONFIG, common mocks for docker-manager, validateOptions) should move to a src/commands/main-action-test-helpers.ts module imported by all four files.

Affected Callers

Test-only file; no production imports. src/cli.ts imports createMainAction from ./commands/main-action (the file under test), which is unaffected by this split.

Effort Estimate

Medium

Benefits

• Each suite becomes independently reviewable and faster to navigate for its concern (dispatch vs. lifecycle vs. cleanup vs. audit)
• Reduces risk of merge conflicts when multiple engineers touch different behaviors of createMainAction
• Shared fixtures/mocks become explicit and reusable rather than re-declared implicitly across unrelated blocks

---

Detected by Refactoring Scanner workflow.

Generated by Refactoring Opportunity Scanner · copilot · auto · 74.7 AIC · ⊞ 10.8K · ◷

• expires on Nov 26, 2026, 3:06 PM UTC

[copilot-swe-agent]

The agent encountered an error and was unable to start working on this issue: This may be caused by a repository ruleset violation. See granting bypass permissions for the agent, or please contact support if the issue persists. (Request id: EBA8:E3A8A:7D1BB51:8298D69:6AB93D53)