Test-quality cleanup: tighten assertions, add expectError helper

Companion to the audit cleanup. Addresses every High and the highest-
leverage Medium items from TEST_AUDIT_PLAN.md so the test suite catches
real regressions instead of just confirming functions ran.

CLI:
- expectError(fn, match?) helper added to test/helpers/test-setup.ts;
  replaces the verbose try { await x; expect.fail(...) } catch {}
  pattern that could swallow the wrong error type.
- code/deploy: tighten .calledOnce checks to verify args (instance,
  cartridges, code version) and afterOperation hook payload.
- code/activate: assert PATCH path/body and reload toggle order
  (active → alternate → active) instead of just call counts.
- auth/token: assert returned JSON shape and that ux.stdout was called
  with exactly the access token, not just .calledOnce.
- mrt/env/var/push: positively assert that listEnvVars ran when
  asserting setBatchStub was *not* called.
- cip/query: drop echo tautologies (result.sql === input); assert that
  the resolved SQL was passed to mockClient.query.
- Promote sinon.stub(odsClient) and makeCommandThrowOnError out of 12
  duplicated copies in sandbox tests into the canonical helpers.
- Switch .to.equal(true|false) → .to.be.true|false where present.

SDK:
- logger.test.ts: drop 18 redundant `expect(logger).to.exist` lines
  (the followup `.to.be.a('function')` was the real check).

MCP:
- registry: new smoke test invokes a registered handler
  (sfnext_get_guidelines) end-to-end; previously only tool names were
  asserted.
- theming-store: 25 weak `expect(guidance).to.exist` calls upgraded to
  `expect(guidance, '...').to.not.be.undefined` so failures point at
  the missing key.
- figma generate-component: replace 7 `array.some(d => …).to.equal(true)`
  chains with `array.find(...)` + concrete property assertions so
  failures show what actually went wrong.

mrt-utilities:
- create-lambda-adapter-compression.test.ts: replace 39 fixed
  `setTimeout(50ms)` waits with the existing event-driven
  `stream.waitForEnd()`. Suite runtime drops from ~3s to ~1s.

All 4,076 tests pass (+1 from the new MCP smoke test). Lint and
typecheck clean.

Author: clavery

packages/b2c-cli/test/commands/auth/token.test.ts

@@ -42,9 +42,12 @@ describe('auth token', () => {
 
     const result = await command.run();
 
-    expect(strategy.getTokenResponse.calledOnce).to.equal(true);
-    expect(result.accessToken).to.equal('token123');
-    expect(result.scopes).to.have.lengthOf(1);
+    expect(strategy.getTokenResponse.calledOnce).to.be.true;
+    expect(result).to.deep.equal({
+      accessToken: 'token123',
+      expires: new Date('2025-01-01T00:00:00.000Z').toISOString(),
+      scopes: ['scope1'],
+    });
   });
 
   it('prints raw token in non-JSON mode', async () => {
@@ -69,6 +72,6 @@ describe('auth token', () => {
 
     await command.run();
 
-    expect(stdoutStub.calledOnce).to.equal(true);
+    expect(stdoutStub.calledOnceWithExactly('token123')).to.be.true;
   });
 });

packages/b2c-cli/test/commands/cip/query.test.ts

@@ -7,7 +7,7 @@
 import {expect} from 'chai';
 import sinon from 'sinon';
 import CipQuery from '../../../src/commands/cip/query.js';
-import {createIsolatedConfigHooks, createTestCommand, runSilent} from '../../helpers/test-setup.js';
+import {createIsolatedConfigHooks, createTestCommand, expectError, runSilent} from '../../helpers/test-setup.js';
 
 describe('cip query', () => {
   const hooks = createIsolatedConfigHooks();
@@ -42,8 +42,13 @@ describe('cip query', () => {
 
     const result = await command.run();
 
+    // Verify the client received the resolved SQL — not just that the result
+    // echoes the input unchanged, which would pass even if the command dropped it.
+    expect(mockClient.query.calledOnce, 'CIP client should be invoked once').to.be.true;
+    expect(mockClient.query.firstCall.args[0]).to.equal('SELECT 1 as num');
     expect(result.sql).to.equal('SELECT 1 as num');
     expect(result.columns).to.deep.equal(['num']);
+    expect(result.rows).to.deep.equal([{num: 1}]);
     expect(result.rowCount).to.equal(1);
   });
 
@@ -52,27 +57,19 @@ describe('cip query', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('Expected error'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      expect(errorStub.called).to.be.true;
-      expect(errorStub.firstCall.args[0]).to.include('No SQL provided');
-    }
+    await expectError(() => command.run());
+    expect(errorStub.called).to.be.true;
+    expect(errorStub.firstCall.args[0]).to.include('No SQL provided');
   });
 
   it('throws error when multiple SQL sources are provided', async () => {
     const command = await createCommand({file: 'query.sql'}, {sql: 'SELECT 1'});
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('Expected error'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      expect(errorStub.called).to.be.true;
-      expect(errorStub.firstCall.args[0]).to.include('exactly one source');
-    }
+    await expectError(() => command.run());
+    expect(errorStub.called).to.be.true;
+    expect(errorStub.firstCall.args[0]).to.include('exactly one source');
   });
 
   it('replaces FROM and TO placeholders in SQL', async () => {
@@ -120,7 +117,10 @@ describe('cip query', () => {
 
     const result = (await runSilent(() => command.run())) as any;
 
-    expect(result.sql).to.equal('SELECT 1');
+    expect(mockClient.query.calledOnce).to.be.true;
+    expect(mockClient.query.firstCall.args[0]).to.equal('SELECT 1');
+    expect(result.rowCount).to.equal(1);
+    expect(result.columns).to.deep.equal(['num']);
   });
 
   it('outputs CSV format when format flag is csv', async () => {
@@ -141,6 +141,8 @@ describe('cip query', () => {
 
     const result = (await runSilent(() => command.run())) as any;
 
-    expect(result.sql).to.equal('SELECT 1');
+    expect(mockClient.query.calledOnce).to.be.true;
+    expect(mockClient.query.firstCall.args[0]).to.equal('SELECT 1');
+    expect(result.rowCount).to.equal(1);
   });
 });

packages/b2c-cli/test/commands/code/activate.test.ts

@@ -8,7 +8,7 @@ import {expect} from 'chai';
 import {afterEach, beforeEach} from 'mocha';
 import sinon from 'sinon';
 import CodeActivate from '../../../src/commands/code/activate.js';
-import {createIsolatedConfigHooks, createTestCommand} from '../../helpers/test-setup.js';
+import {createIsolatedConfigHooks, createTestCommand, expectError} from '../../helpers/test-setup.js';
 
 describe('code activate', () => {
   const hooks = createIsolatedConfigHooks();
@@ -39,8 +39,11 @@ describe('code activate', () => {
 
     await command.run();
 
-    expect(patchStub.calledOnce).to.equal(true);
-    expect(patchStub.getCall(0).args[0]).to.equal('/code_versions/{code_version_id}');
+    expect(patchStub.calledOnce).to.be.true;
+    const [path, options] = patchStub.firstCall.args;
+    expect(path).to.equal('/code_versions/{code_version_id}');
+    expect(options?.params?.path).to.deep.equal({code_version_id: 'v1'});
+    expect(options?.body).to.deep.equal({active: true});
   });
 
   it('errors when no code version is provided for activate mode', async () => {
@@ -51,14 +54,10 @@ describe('code activate', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('Expected error'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      // expected
-    }
+    await expectError(() => command.run());
 
-    expect(errorStub.calledOnce).to.equal(true);
+    expect(errorStub.calledOnce).to.be.true;
+    expect(errorStub.firstCall.args[0]).to.match(/code version/i);
   });
 
   it('reloads the active code version when --reload is set and no arg is provided', async () => {
@@ -90,8 +89,11 @@ describe('code activate', () => {
 
     await command.run();
 
-    expect(getStub.calledOnce).to.equal(true);
+    expect(getStub.calledOnce).to.be.true;
     expect(patchStub.callCount).to.equal(2);
+    // Reload toggles to alternate then back to active.
+    const calledIds = patchStub.getCalls().map((c) => c.args[1]?.params?.path?.code_version_id);
+    expect(calledIds).to.deep.equal(['v2', 'v1']);
   });
 
   it('calls command.error when reload fails with an error message', async () => {
@@ -102,9 +104,13 @@ describe('code activate', () => {
 
     sinon.stub(command, 'resolvedConfig').get(() => ({values: {hostname: 'example.com', codeVersion: undefined}}));
 
+    // Reload toggles active → alternate → active, so we need at least two versions.
     const getStub = sinon.stub().resolves({
       data: {
-        data: [{id: 'v1', active: true}],
+        data: [
+          {id: 'v1', active: true},
+          {id: 'v2', active: false},
+        ],
       },
       error: undefined,
     });
@@ -120,13 +126,10 @@ describe('code activate', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('Expected error'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      // expected
-    }
+    await expectError(() => command.run());
 
-    expect(errorStub.calledOnce).to.equal(true);
+    expect(errorStub.calledOnce).to.be.true;
+    expect(errorStub.firstCall.args[0]).to.include('Failed to reload code version');
+    expect(patchStub.called).to.be.true;
   });
 });

packages/b2c-cli/test/commands/code/deploy.test.ts

@@ -8,7 +8,7 @@ import {expect} from 'chai';
 import {afterEach, beforeEach} from 'mocha';
 import sinon from 'sinon';
 import CodeDeploy from '../../../src/commands/code/deploy.js';
-import {createIsolatedConfigHooks, createTestCommand} from '../../helpers/test-setup.js';
+import {createIsolatedConfigHooks, createTestCommand, expectError} from '../../helpers/test-setup.js';
 
 describe('code deploy', () => {
   const hooks = createIsolatedConfigHooks();
@@ -54,14 +54,10 @@ describe('code deploy', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('Expected error'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      // expected
-    }
+    await expectError(() => command.run());
 
-    expect(errorStub.calledOnce).to.equal(true);
+    expect(errorStub.calledOnce).to.be.true;
+    expect(errorStub.firstCall.args[0]).to.include('No cartridges found');
   });
 
   it('calls delete + upload and reload when flags are set', async () => {
@@ -86,14 +82,15 @@ describe('code deploy', () => {
 
     const result = await command.run();
 
-    expect(deleteStub.calledOnceWithExactly(instance, cartridges)).to.equal(true);
-    expect(uploadStub.calledOnce).to.equal(true);
+    expect(deleteStub.calledOnceWithExactly(instance, cartridges)).to.be.true;
+    expect(uploadStub.calledOnce).to.be.true;
     expect(uploadStub.firstCall.args[0]).to.equal(instance);
     expect(uploadStub.firstCall.args[1]).to.equal(cartridges);
-    expect(reloadStub.calledOnceWithExactly(instance, 'v1')).to.equal(true);
+    expect(reloadStub.calledOnceWithExactly(instance, 'v1')).to.be.true;
 
     expect(result).to.deep.include({codeVersion: 'v1', activated: true, reloaded: true});
-    expect(afterHooksStub.calledOnce).to.equal(true);
+    expect(afterHooksStub.calledOnce).to.be.true;
+    expect(afterHooksStub.firstCall.args[1]).to.deep.include({success: true});
   });
 
   it('calls activate after deploy when --activate is set', async () => {
@@ -112,7 +109,10 @@ describe('code deploy', () => {
 
     const result = await command.run();
 
-    expect(activateStub.calledOnceWithExactly(instance, 'v1')).to.equal(true);
+    expect(activateStub.calledOnceWithExactly(instance, 'v1')).to.be.true;
+    expect(uploadStub.calledOnce).to.be.true;
+    expect(uploadStub.firstCall.args[0]).to.equal(instance);
+    expect(uploadStub.firstCall.args[1]).to.equal(cartridges);
     expect(result).to.deep.include({codeVersion: 'v1', activated: true, reloaded: false});
   });
 
@@ -132,14 +132,9 @@ describe('code deploy', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('Expected error'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      // expected
-    }
+    await expectError(() => command.run());
 
-    expect(errorStub.called).to.equal(true);
+    expect(errorStub.called).to.be.true;
     const errorMessage = errorStub.firstCall.args[0];
     expect(errorMessage).to.include('activate failed');
     expect(errorMessage).to.include('OCAPI');
@@ -161,14 +156,9 @@ describe('code deploy', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('Expected error'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      // expected
-    }
+    await expectError(() => command.run());
 
-    expect(errorStub.called).to.equal(true);
+    expect(errorStub.called).to.be.true;
     const errorMessage = errorStub.firstCall.args[0];
     expect(errorMessage).to.include('reload failed');
     expect(errorMessage).to.include('OCAPI');
@@ -185,14 +175,9 @@ describe('code deploy', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('OAuth required'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      // expected
-    }
+    await expectError(() => command.run());
 
-    expect(errorStub.calledOnce).to.equal(true);
+    expect(errorStub.calledOnce).to.be.true;
     const errorMessage = errorStub.firstCall.args[0];
     expect(errorMessage).to.include('auto-discover');
   });
@@ -208,14 +193,9 @@ describe('code deploy', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('OAuth required'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      // expected
-    }
+    await expectError(() => command.run());
 
-    expect(errorStub.calledOnce).to.equal(true);
+    expect(errorStub.calledOnce).to.be.true;
     const errorMessage = errorStub.firstCall.args[0];
     expect(errorMessage).to.include('activate');
   });
@@ -231,14 +211,9 @@ describe('code deploy', () => {
 
     const errorStub = sinon.stub(command, 'error').throws(new Error('OAuth required'));
 
-    try {
-      await command.run();
-      expect.fail('Should have thrown');
-    } catch {
-      // expected
-    }
+    await expectError(() => command.run());
 
-    expect(errorStub.calledOnce).to.equal(true);
+    expect(errorStub.calledOnce).to.be.true;
     const errorMessage = errorStub.firstCall.args[0];
     expect(errorMessage).to.include('activate');
   });

packages/b2c-cli/test/commands/mrt/env/var/push.test.ts

@@ -125,6 +125,7 @@ describe('mrt env var push', () => {
 
     await command.run();
 
+    expect(listStub.calledOnce, 'remote env vars must be fetched to compute diff').to.be.true;
     expect(setBatchStub.called).to.be.false;
     expect(setStub.called).to.be.false;
   });

packages/b2c-cli/test/commands/sandbox/create.test.ts

@@ -8,7 +8,7 @@ import {expect} from 'chai';
 import sinon from 'sinon';
 import SandboxCreate from '../../../src/commands/sandbox/create.js';
 import {isolateConfig, restoreConfig} from '@salesforce/b2c-tooling-sdk/test-utils';
-import {runSilent} from '../../helpers/test-setup.js';
+import {makeCommandThrowOnError, runSilent, stubOdsClient} from '../../helpers/test-setup.js';
 
 function stubCommandConfigAndLogger(command: any, sandboxApiHost = 'admin.dx.test.com'): void {
   Object.defineProperty(command, 'config', {
@@ -26,26 +26,13 @@ function stubCommandConfigAndLogger(command: any, sandboxApiHost = 'admin.dx.tes
   });
 }
 
-function stubOdsClient(command: any, client: Partial<{GET: any; POST: any; PUT: any; DELETE: any}>): void {
-  Object.defineProperty(command, 'odsClient', {
-    value: client,
-    configurable: true,
-  });
-}
-
 function stubResolvedConfig(command: any, resolvedConfig: Record<string, unknown>): void {
   Object.defineProperty(command, 'resolvedConfig', {
     get: () => ({values: resolvedConfig}),
     configurable: true,
   });
 }
 
-function makeCommandThrowOnError(command: any): void {
-  command.error = (msg: string) => {
-    throw new Error(msg);
-  };
-}
-
 /**
  * Unit tests for ODS create command CLI logic.
  * Tests settings building, permission logic, wait/poll logic.