test: C2c — lock the host, exit and control behavior; delete pointless tests

src/agent/__tests__/__snapshots__/credential-isolation-snapshot.test.ts.snap

@@ -1,20 +1,5 @@
 // Vitest Snapshot v1, https://vitest.dev/guide/snapshot.html
 
-exports[`credential-isolation paths (snapshot) > settings-file conflicts: neutralize / warn / fail-closed disposition 1`] = `
-{
-  "autoFix_neutralize": [
-    "project",
-  ],
-  "failClosed_userFixes": [
-    "managed",
-  ],
-  "warnOnly_alreadyIgnored": [
-    "user",
-    "project-local",
-  ],
-}
-`;
-
 exports[`credential-isolation paths (snapshot) > subprocess env: disposition of every credential path 1`] = `
 {
   "ANTHROPIC_API_KEY": "stripped",

src/agent/__tests__/commandments.test.ts

@@ -2,8 +2,6 @@ import { WIZARD_COMMANDMENTS } from '@agent/commandments';
 import { assembleCommandments } from '@agent/runner/switchboard/commandments';
 import { Harness, Sequence } from '@shared/constants';
 
-const global = WIZARD_COMMANDMENTS.join('\n');
-
 /** Every axis combination that reaches a runner today. */
 const CAPS = { bash: true, posthogMcp: true };
 const prompt = (
@@ -55,25 +53,6 @@ describe('commandments by axis', () => {
         );
       },
     );
-
-    it('are stated once, not once per source', () => {
-      // Previously duplicated: the global list said "create tasks as soon as you
-      // understand the work" while pi's runtime notes said "after you load and
-      // skim the skill workflow, not before" — both in the same prompt.
-      const linear = prompt(Harness.pi, Sequence.linear);
-      expect(
-        linear.match(/Create the task list once you understand/g),
-      ).toHaveLength(1);
-      expect(linear.match(/Each task subject is SHORT/g)).toHaveLength(1);
-      expect(linear).not.toMatch(/Create tasks as soon as you understand/);
-    });
-
-    it('keeps the provider-naming rule global — it governs code, not a tool', () => {
-      expect(global).toMatch(/Do not assume "PostHog provider"/);
-      expect(prompt(Harness.pi, Sequence.orchestrator)).toMatch(
-        /Do not assume "PostHog provider"/,
-      );
-    });
   });
 
   describe('axis scoping', () => {
@@ -99,41 +78,6 @@ describe('commandments by axis', () => {
       );
     });
   });
-
-  // Targeted assertions for the wizard_ask Path A translation rules.
-  // These are the rules a skill author depends on when leaving their prose
-  // unchanged — they need to keep working as the commandment list evolves.
-  describe('wizard_ask Path A rules', () => {
-    const text = global;
-
-    it('names the tool explicitly', () => {
-      expect(text).toMatch(/`wizard_ask`/);
-    });
-
-    it('forbids inlining questions in text output', () => {
-      expect(text).toMatch(/never inline questions/i);
-    });
-
-    it('requires batching prose lists into one call', () => {
-      expect(text).toMatch(/single `wizard_ask` tool call/i);
-      expect(text).toMatch(/never split/i);
-    });
-
-    it('describes how to infer `kind`', () => {
-      expect(text).toMatch(/`single`/);
-      expect(text).toMatch(/`multi`/);
-      expect(text).toMatch(/`text`/);
-    });
-
-    it('describes how to derive options and ids', () => {
-      expect(text).toMatch(/kebab-case/i);
-      expect(text).toMatch(/label.*value/i);
-    });
-
-    it('tells the agent to use answers directly without re-asking', () => {
-      expect(text).toMatch(/do not re-ask/i);
-    });
-  });
 });
 
 /**

src/agent/__tests__/credential-isolation-snapshot.test.ts

@@ -14,8 +14,6 @@
  */
 
 import { sanitizeAgentSubprocessEnv } from '@shared/agent-env-isolation';
-import { classifySettingsConflicts } from '@shared/claude-settings';
-import type { SettingsConflict } from '@shared/claude-settings';
 
 // Every env-based avenue (one key each), plus the gateway routing and benign
 // env that must survive. Grouped by avenue for readability; the snapshot sorts.
@@ -91,46 +89,4 @@ describe('credential-isolation paths (snapshot)', () => {
     );
     expect(disposition).toMatchSnapshot();
   });
-
-  it('settings-file conflicts: neutralize / warn / fail-closed disposition', () => {
-    const conflicts: SettingsConflict[] = [
-      // writable project file — SDK reads it; wizard can remove it
-      {
-        source: 'project',
-        path: '/proj/.claude/settings.json',
-        keys: ['ANTHROPIC_BASE_URL'],
-        writable: true,
-      },
-      // org-managed — always read, unremovable
-      {
-        source: 'managed',
-        path: '/managed/managed-settings.json',
-        keys: ['apiKeyHelper'],
-        writable: false,
-      },
-      // user global — ignored under settingSources:['project']
-      {
-        source: 'user',
-        path: '/home/.claude/settings.json',
-        keys: ['ANTHROPIC_API_KEY'],
-        writable: false,
-      },
-      // project-local — ignored under settingSources:['project']
-      {
-        source: 'project-local',
-        path: '/proj/.claude/settings.local.json',
-        keys: ['CLAUDE_CODE_USE_BEDROCK'],
-        writable: false,
-      },
-    ];
-
-    const { autoFix, failClosed, warnOnly } =
-      classifySettingsConflicts(conflicts);
-
-    expect({
-      autoFix_neutralize: autoFix.map((c) => c.source),
-      failClosed_userFixes: failClosed.map((c) => c.source),
-      warnOnly_alreadyIgnored: warnOnly.map((c) => c.source),
-    }).toMatchSnapshot();
-  });
 });