fix(macp): fail-closed typed gate states — no placeholder/empty-command passes (#1275)

RI-N2 / SDLC-D-035. Every GateResult now carries a typed status
discriminator (passed|failed|simulated|waiting|capability_failure);
passed:true remains true only for really-executed, really-green gates.

- ci-pipeline without a CI provider → capability_failure (MACP_NO_CI_PIPELINE),
  never the placeholder pass
- empty-command gate → capability_failure (MACP_NO_COMMAND / MACP_NO_REVIEWER),
  and runGates no longer silently skips it
- manual gate type waits (MACP_AUTHORITY_REQUIRED) — neither pass nor fail
- explicit simulation only (simulate option / --simulate): typed simulated
  results can never make the aggregate passed (RunGatesResult.state)
- CLI stubs (tasks list, submit, events tail) exit nonzero with typed
  MACP_NOT_IMPLEMENTED; macp gate is now wired to runGates with --simulate
- legacy __tests__/gate-runner.test.ts consolidated into src/gate-runner.spec.ts
  (root eslint project service does not cover packages/macp/__tests__)

packages/macp/__tests__/gate-runner.test.ts

@@ -1,253 +0,0 @@
-import { mkdirSync, readFileSync, rmSync } from 'node:fs';
-import { join } from 'node:path';
-import { tmpdir } from 'node:os';
-import { randomUUID } from 'node:crypto';
-import { describe, it, expect, beforeEach, afterEach } from 'vitest';
-import { normalizeGate, countAIFindings, runGate, runGates } from '../src/gate-runner.js';
-
-function makeTmpDir(): string {
-  const dir = join(tmpdir(), `macp-gate-${randomUUID()}`);
-  mkdirSync(dir, { recursive: true });
-  return dir;
-}
-
-describe('normalizeGate', () => {
-  it('normalizes a string to mechanical gate', () => {
-    expect(normalizeGate('echo test')).toEqual({
-      command: 'echo test',
-      type: 'mechanical',
-      fail_on: 'blocker',
-    });
-  });
-
-  it('normalizes an object gate with defaults', () => {
-    expect(normalizeGate({ command: 'lint' })).toEqual({
-      command: 'lint',
-      type: 'mechanical',
-      fail_on: 'blocker',
-    });
-  });
-
-  it('preserves explicit type and fail_on', () => {
-    expect(normalizeGate({ command: 'review', type: 'ai-review', fail_on: 'any' })).toEqual({
-      command: 'review',
-      type: 'ai-review',
-      fail_on: 'any',
-    });
-  });
-
-  it('handles non-string/non-object input', () => {
-    expect(normalizeGate(42)).toEqual({ command: '', type: 'mechanical', fail_on: 'blocker' });
-    expect(normalizeGate(null)).toEqual({ command: '', type: 'mechanical', fail_on: 'blocker' });
-  });
-});
-
-describe('countAIFindings', () => {
-  it('returns zeros for non-object', () => {
-    expect(countAIFindings(null)).toEqual({ blockers: 0, total: 0 });
-    expect(countAIFindings('string')).toEqual({ blockers: 0, total: 0 });
-    expect(countAIFindings([])).toEqual({ blockers: 0, total: 0 });
-  });
-
-  it('counts from stats block', () => {
-    const output = { stats: { blockers: 2, should_fix: 3, suggestions: 1 } };
-    expect(countAIFindings(output)).toEqual({ blockers: 2, total: 6 });
-  });
-
-  it('counts from findings array when stats has no blockers', () => {
-    const output = {
-      stats: { blockers: 0 },
-      findings: [{ severity: 'blocker' }, { severity: 'warning' }, { severity: 'blocker' }],
-    };
-    expect(countAIFindings(output)).toEqual({ blockers: 2, total: 3 });
-  });
-
-  it('uses stats blockers over findings array when stats has blockers', () => {
-    const output = {
-      stats: { blockers: 5 },
-      findings: [{ severity: 'blocker' }, { severity: 'warning' }],
-    };
-    // stats.blockers = 5, total from stats = 5+0+0 = 5, findings not used for total since stats total is non-zero
-    expect(countAIFindings(output)).toEqual({ blockers: 5, total: 5 });
-  });
-
-  it('counts findings length as total when stats has zero total', () => {
-    const output = {
-      findings: [{ severity: 'warning' }, { severity: 'info' }],
-    };
-    expect(countAIFindings(output)).toEqual({ blockers: 0, total: 2 });
-  });
-});
-
-describe('runGate', () => {
-  let tmp: string;
-  let logPath: string;
-
-  beforeEach(() => {
-    tmp = makeTmpDir();
-    logPath = join(tmp, 'gate.log');
-  });
-
-  afterEach(() => {
-    rmSync(tmp, { recursive: true, force: true });
-  });
-
-  it('passes mechanical gate on exit 0', () => {
-    const result = runGate('echo hello', tmp, logPath, 30);
-    expect(result.passed).toBe(true);
-    expect(result.exit_code).toBe(0);
-    expect(result.type).toBe('mechanical');
-    expect(result.output).toContain('hello');
-  });
-
-  it('fails mechanical gate on non-zero exit', () => {
-    const result = runGate('exit 1', tmp, logPath, 30);
-    expect(result.passed).toBe(false);
-    expect(result.exit_code).toBe(1);
-  });
-
-  it('ci-pipeline always passes', () => {
-    const result = runGate({ command: 'anything', type: 'ci-pipeline' }, tmp, logPath, 30);
-    expect(result.passed).toBe(true);
-    expect(result.type).toBe('ci-pipeline');
-    expect(result.output).toBe('CI pipeline gate placeholder');
-  });
-
-  it('empty command passes', () => {
-    const result = runGate({ command: '' }, tmp, logPath, 30);
-    expect(result.passed).toBe(true);
-  });
-
-  it('ai-review gate parses JSON output', () => {
-    const json = JSON.stringify({ stats: { blockers: 0, should_fix: 1 } });
-    const result = runGate({ command: `echo '${json}'`, type: 'ai-review' }, tmp, logPath, 30);
-    expect(result.passed).toBe(true);
-    expect(result.blockers).toBe(0);
-    expect(result.findings).toBe(1);
-  });
-
-  it('ai-review gate fails on blockers', () => {
-    const json = JSON.stringify({ stats: { blockers: 2 } });
-    const result = runGate({ command: `echo '${json}'`, type: 'ai-review' }, tmp, logPath, 30);
-    expect(result.passed).toBe(false);
-    expect(result.blockers).toBe(2);
-  });
-
-  it('ai-review gate with fail_on=any fails on any findings', () => {
-    const json = JSON.stringify({ stats: { blockers: 0, should_fix: 1 } });
-    const result = runGate(
-      { command: `echo '${json}'`, type: 'ai-review', fail_on: 'any' },
-      tmp,
-      logPath,
-      30,
-    );
-    expect(result.passed).toBe(false);
-    expect(result.fail_on).toBe('any');
-  });
-
-  it('ai-review gate fails on invalid JSON output', () => {
-    const result = runGate({ command: 'echo "not json"', type: 'ai-review' }, tmp, logPath, 30);
-    expect(result.passed).toBe(false);
-    expect(result.parse_error).toBeDefined();
-  });
-
-  it('writes to log file', () => {
-    runGate('echo logged', tmp, logPath, 30);
-    const log = readFileSync(logPath, 'utf-8');
-    expect(log).toContain('COMMAND: echo logged');
-    expect(log).toContain('logged');
-    expect(log).toContain('EXIT:');
-  });
-});
-
-describe('runGates', () => {
-  let tmp: string;
-  let logPath: string;
-  let eventsPath: string;
-
-  beforeEach(() => {
-    tmp = makeTmpDir();
-    logPath = join(tmp, 'gates.log');
-    eventsPath = join(tmp, 'events.ndjson');
-  });
-
-  afterEach(() => {
-    rmSync(tmp, { recursive: true, force: true });
-  });
-
-  it('runs multiple gates and returns results', () => {
-    const { allPassed, gateResults } = runGates(
-      ['echo one', 'echo two'],
-      tmp,
-      logPath,
-      30,
-      eventsPath,
-      'task-1',
-    );
-    expect(allPassed).toBe(true);
-    expect(gateResults).toHaveLength(2);
-  });
-
-  it('reports failure when any gate fails', () => {
-    const { allPassed, gateResults } = runGates(
-      ['echo ok', 'exit 1'],
-      tmp,
-      logPath,
-      30,
-      eventsPath,
-      'task-2',
-    );
-    expect(allPassed).toBe(false);
-    expect(gateResults[0]!.passed).toBe(true);
-    expect(gateResults[1]!.passed).toBe(false);
-  });
-
-  it('emits events for each gate', () => {
-    runGates(['echo test'], tmp, logPath, 30, eventsPath, 'task-3');
-    const events = readFileSync(eventsPath, 'utf-8')
-      .trim()
-      .split('\n')
-      .map((l) => JSON.parse(l));
-    expect(events).toHaveLength(2); // started + passed
-    expect(events[0].event_type).toBe('rail.check.started');
-    expect(events[1].event_type).toBe('rail.check.passed');
-  });
-
-  it('skips gates with empty command (non ci-pipeline)', () => {
-    const { gateResults } = runGates(
-      [{ command: '', type: 'mechanical' }, 'echo real'],
-      tmp,
-      logPath,
-      30,
-      eventsPath,
-      'task-4',
-    );
-    expect(gateResults).toHaveLength(1);
-  });
-
-  it('does not skip ci-pipeline even with empty command', () => {
-    const { gateResults } = runGates(
-      [{ command: '', type: 'ci-pipeline' }],
-      tmp,
-      logPath,
-      30,
-      eventsPath,
-      'task-5',
-    );
-    expect(gateResults).toHaveLength(1);
-    expect(gateResults[0]!.passed).toBe(true);
-  });
-
-  it('emits failed event with correct message', () => {
-    runGates(['exit 42'], tmp, logPath, 30, eventsPath, 'task-6');
-    const events = readFileSync(eventsPath, 'utf-8')
-      .trim()
-      .split('\n')
-      .map((l) => JSON.parse(l));
-    const failEvent = events.find(
-      (e: Record<string, unknown>) => e.event_type === 'rail.check.failed',
-    );
-    expect(failEvent).toBeDefined();
-    expect(failEvent.message).toContain('Gate failed (');
-  });
-});

packages/macp/src/cli.spec.ts

@@ -1,5 +1,8 @@
-import { describe, it, expect } from 'vitest';
+import { describe, it, expect, afterEach, beforeEach, vi } from 'vitest';
 import { Command } from 'commander';
+import fs from 'node:fs';
+import os from 'node:os';
+import path from 'node:path';
 import { registerMacpCommand } from './cli.js';
 
 describe('registerMacpCommand', () => {
@@ -75,3 +78,162 @@ describe('registerMacpCommand', () => {
     expect(topLevel).toContain('events');
   });
 });
+
+/**
+ * RI-N2 fail-closed CLI behavior: an unimplemented capability is a failure,
+ * never a success. Every stub exits nonzero with a typed message, and the
+ * implemented `macp gate` mirrors the typed gate-runner states.
+ */
+describe('registerMacpCommand fail-closed (RI-N2)', () => {
+  let tmpDir: string;
+
+  function buildProgram(): Command {
+    const program = new Command();
+    program.exitOverride();
+    program.configureOutput({ writeErr: () => {} });
+    registerMacpCommand(program);
+    return program;
+  }
+
+  beforeEach(() => {
+    tmpDir = fs.mkdtempSync(path.join(os.tmpdir(), 'macp-cli-failclosed-'));
+    process.exitCode = 0;
+  });
+
+  afterEach(() => {
+    process.exitCode = 0;
+    fs.rmSync(tmpDir, { recursive: true, force: true });
+  });
+
+  it('macp tasks list exits nonzero (unimplemented capability)', async () => {
+    const program = buildProgram();
+    await program.parseAsync(['macp', 'tasks', 'list'], { from: 'user' });
+    expect(process.exitCode).not.toBe(0);
+  });
+
+  it('macp submit exits nonzero with a typed MACP_NOT_IMPLEMENTED message', async () => {
+    const program = buildProgram();
+    const errSpy = vi.spyOn(console, 'error').mockImplementation(() => {});
+    try {
+      await program.parseAsync(['macp', 'submit', 'spec.json'], { from: 'user' });
+      expect(process.exitCode).not.toBe(0);
+      const errText = errSpy.mock.calls.map((c) => String(c[0])).join('\n');
+      expect(errText).toContain('MACP_NOT_IMPLEMENTED');
+    } finally {
+      errSpy.mockRestore();
+    }
+  });
+
+  it('macp events tail exits nonzero (unimplemented capability)', async () => {
+    const program = buildProgram();
+    await program.parseAsync(['macp', 'events', 'tail'], { from: 'user' });
+    expect(process.exitCode).not.toBe(0);
+  });
+
+  it('macp gate runs a green inline command and exits 0', async () => {
+    const program = buildProgram();
+    await program.parseAsync(
+      [
+        'macp',
+        'gate',
+        'exit 0',
+        '--cwd',
+        tmpDir,
+        '--log',
+        path.join(tmpDir, 'g.log'),
+        '--timeout',
+        '10',
+      ],
+      { from: 'user' },
+    );
+    expect(process.exitCode).toBe(0);
+  });
+
+  it('macp gate exits nonzero on a failing command', async () => {
+    const program = buildProgram();
+    await program.parseAsync(
+      [
+        'macp',
+        'gate',
+        'exit 9',
+        '--cwd',
+        tmpDir,
+        '--log',
+        path.join(tmpDir, 'g.log'),
+        '--timeout',
+        '10',
+      ],
+      { from: 'user' },
+    );
+    expect(process.exitCode).not.toBe(0);
+  });
+
+  it('macp gate with an unimplemented ci-pipeline capability exits nonzero', async () => {
+    const program = buildProgram();
+    const specPath = path.join(tmpDir, 'gates.json');
+    fs.writeFileSync(specPath, JSON.stringify([{ type: 'ci-pipeline' }]));
+    await program.parseAsync(
+      [
+        'macp',
+        'gate',
+        specPath,
+        '--cwd',
+        tmpDir,
+        '--log',
+        path.join(tmpDir, 'g.log'),
+        '--timeout',
+        '10',
+      ],
+      { from: 'user' },
+    );
+    expect(process.exitCode).not.toBe(0);
+  });
+
+  it('macp gate --simulate completes (exit 0) but reports simulated results', async () => {
+    const program = buildProgram();
+    const logSpy = vi.spyOn(console, 'log').mockImplementation(() => {});
+    try {
+      await program.parseAsync(
+        [
+          'macp',
+          'gate',
+          'exit 0',
+          '--simulate',
+          '--cwd',
+          tmpDir,
+          '--log',
+          path.join(tmpDir, 'g.log'),
+          '--timeout',
+          '10',
+        ],
+        { from: 'user' },
+      );
+      // completes only because the caller explicitly asked to simulate
+      expect(process.exitCode).toBe(0);
+      const outText = logSpy.mock.calls.map((c) => String(c[0])).join('\n');
+      expect(outText).toContain('simulated');
+      expect(outText).toContain('SIMULATED');
+    } finally {
+      logSpy.mockRestore();
+    }
+  });
+
+  it('macp gate with an empty spec exits nonzero with a typed error', async () => {
+    const program = buildProgram();
+    await program.parseAsync(
+      [
+        'macp',
+        'gate',
+        '   ',
+        '--cwd',
+        tmpDir,
+        '--log',
+        path.join(tmpDir, 'g.log'),
+        '--timeout',
+        '10',
+      ],
+      { from: 'user' },
+    );
+    expect(process.exitCode).not.toBe(0);
+  });
+});

packages/macp/src/cli.ts

@@ -1,5 +1,73 @@
+import { existsSync, readFileSync } from 'node:fs';
+
 import type { Command } from 'commander';
 
+import { runGates } from './gate-runner.js';
+import { MACPCapabilityError, type MacpErrorCode } from './errors.js';
+
+/**
+ * Load gates from a spec: an existing file (JSON gates array, a JSON object
+ * with `quality_gates`, a JSON gate object, or one command per line) or an
+ * inline command string. Fails closed with a typed capability error when the
+ * spec contains no executable gate definition.
+ */
+function loadGateSpec(spec: string): unknown[] {
+  if (existsSync(spec)) {
+    const raw = readFileSync(spec, 'utf-8');
+    try {
+      const parsed = JSON.parse(raw) as unknown;
+      if (Array.isArray(parsed)) {
+        if (parsed.length === 0) {
+          throw new MACPCapabilityError(
+            'MACP_NO_COMMAND',
+            'gate-spec',
+            `gate spec file '${spec}' contains an empty gates array`,
+          );
+        }
+        return parsed;
+      }
+      if (typeof parsed === 'object' && parsed !== null) {
+        const obj = parsed as Record<string, unknown>;
+        if (Array.isArray(obj['quality_gates'])) {
+          return obj['quality_gates'];
+        }
+        return [parsed];
+      }
+      throw new MACPCapabilityError(
+        'MACP_NO_COMMAND',
+        'gate-spec',
+        `gate spec file '${spec}' parsed to ${typeof parsed} — expected a gates array, a task with quality_gates, or a gate object`,
+      );
+    } catch (exc) {
+      if (exc instanceof MACPCapabilityError) throw exc;
+      // Not JSON — treat each non-empty line as a command gate.
+      const lines = raw
+        .split('\n')
+        .map((l) => l.trim())
+        .filter((l) => l.length > 0);
+      if (lines.length > 0) return lines;
+      throw new MACPCapabilityError(
+        'MACP_NO_COMMAND',
+        'gate-spec',
+        `gate spec file '${spec}' contains no gates`,
+      );
+    }
+  }
+  if (spec.trim().length > 0) return [spec];
+  throw new MACPCapabilityError('MACP_NO_COMMAND', 'gate-spec', 'gate spec is empty');
+}
+
+/** Print a typed not-implemented failure and exit nonzero (RI-N2 fail-closed). */
+function notImplemented(subcommand: string, capability: string, hint: string): void {
+  const err = new MACPCapabilityError(
+    'MACP_NOT_IMPLEMENTED',
+    capability,
+    `${subcommand} is not implemented in @mosaicstack/macp yet (${capability} capability absent) — ${hint}`,
+  );
+  console.error(`[macp] ${subcommand}: ${err.message} [${err.code}]`);
+  process.exitCode = 1;
+}
+
 /**
  * Register macp subcommands on an existing Commander program.
  * This avoids cross-package Commander version mismatches by using the
@@ -24,15 +92,14 @@ export function registerMacpCommand(parent: Command): void {
       'Filter by task type (coding|deploy|research|review|documentation|infrastructure)',
     )
     .action((opts: { status?: string; type?: string }) => {
-      // not yet wired — task persistence layer is not present in @mosaicstack/macp
-      console.log('[macp] tasks list: not yet wired — use macp package programmatically');
+      // unimplemented capability — a failure, never a success (RI-N2)
       if (opts.status) {
         console.log(`  status filter: ${opts.status}`);
       }
       if (opts.type) {
         console.log(`  type filter: ${opts.type}`);
       }
-      process.exitCode = 0;
+      notImplemented('tasks list', 'task-persistence', 'use the macp package programmatically');
     });
 
   // ─── submit ──────────────────────────────────────────────────────────────
@@ -41,12 +108,11 @@ export function registerMacpCommand(parent: Command): void {
     .command('submit <path>')
     .description('Submit a task from a JSON/YAML spec file')
     .action((specPath: string) => {
-      // not yet wired — task submission requires a running MACP server
-      console.log('[macp] submit: not yet wired — use macp package programmatically');
+      // unimplemented capability — a failure, never a success (RI-N2)
       console.log(`  spec path: ${specPath}`);
       console.log('  task id:   (unavailable — no MACP server connected)');
       console.log('  status:    (unavailable — no MACP server connected)');
-      process.exitCode = 0;
+      notImplemented('submit', 'macp-server', 'use the macp package programmatically');
     });
 
   // ─── gate ────────────────────────────────────────────────────────────────
@@ -58,16 +124,58 @@ export function registerMacpCommand(parent: Command): void {
     .option('--cwd <path>', 'Working directory for gate execution', process.cwd())
     .option('--log <path>', 'Path to write gate log output', '/tmp/macp-gate.log')
     .option('--timeout <seconds>', 'Gate timeout in seconds', '60')
-    .action((spec: string, opts: { failOn: string; cwd: string; log: string; timeout: string }) => {
-      // not yet wired — gate execution requires a task context and event sink
-      console.log('[macp] gate: not yet wired — use macp package programmatically');
-      console.log(`  spec:     ${spec}`);
-      console.log(`  fail-on:  ${opts.failOn}`);
-      console.log(`  cwd:      ${opts.cwd}`);
-      console.log(`  log:      ${opts.log}`);
-      console.log(`  timeout:  ${opts.timeout}s`);
-      process.exitCode = 0;
-    });
+    .option(
+      '--simulate',
+      'Simulate gates instead of executing them; results are typed simulated and never satisfy a check',
+    )
+    .action(
+      (
+        spec: string,
+        opts: { failOn: string; cwd: string; log: string; timeout: string; simulate?: boolean },
+      ) => {
+        let gates: unknown[];
+        try {
+          gates = loadGateSpec(spec);
+        } catch (exc) {
+          if (exc instanceof MACPCapabilityError) {
+            console.error(`[macp] gate: ${exc.message} [${exc.code}]`);
+          } else {
+            console.error(`[macp] gate: ${String(exc)}`);
+          }
+          process.exitCode = 1;
+          return;
+        }
+
+        const timeoutSec = Number.parseInt(opts.timeout, 10) || 60;
+        const eventsPath = `${opts.log}.events.ndjson`;
+        const { state, gateResults } = runGates(
+          gates,
+          opts.cwd,
+          opts.log,
+          timeoutSec,
+          eventsPath,
+          'macp-cli-gate',
+          {
+            simulate: opts.simulate,
+          },
+        );
+
+        for (const r of gateResults) {
+          const label = r.command || r.type;
+          const reason = r.reason ? ` — ${r.reason}` : '';
+          console.log(`[macp] gate ${r.status}: ${label}${reason}`);
+        }
+        if (opts.simulate) {
+          console.log(
+            '[macp] SIMULATED run — every result is typed simulated and can never satisfy a gate, dependency, or release check',
+          );
+        }
+
+        // Simulated runs may complete (exit 0) only because the caller
+        // explicitly passed --simulate; the typed state stays 'simulated'.
+        process.exitCode = state === 'passed' || state === 'simulated' ? 0 : 1;
+      },
+    );
 
   // ─── events ──────────────────────────────────────────────────────────────
 
@@ -79,14 +187,16 @@ export function registerMacpCommand(parent: Command): void {
     .option('--file <path>', 'Path to the MACP events NDJSON file')
     .option('--follow', 'Follow the file for new events (like tail -f)')
     .action((opts: { file?: string; follow?: boolean }) => {
-      // not yet wired — event streaming requires a live event source
-      console.log('[macp] events tail: not yet wired — use macp package programmatically');
+      // unimplemented capability — a failure, never a success (RI-N2)
       if (opts.file) {
         console.log(`  file:   ${opts.file}`);
       }
       if (opts.follow) {
         console.log('  mode:   follow');
       }
-      process.exitCode = 0;
+      notImplemented('events tail', 'event-source', 'use the macp package programmatically');
     });
 }
+
+// Re-export so CLI consumers can surface typed capability codes.
+export type { MacpErrorCode };

packages/macp/src/errors.ts

@@ -0,0 +1,35 @@
+/** Typed error code from the closed MACP_ERROR_CODES set. */
+export type MacpErrorCode = (typeof MACP_ERROR_CODES)[number];
+/**
+ * Typed fail-closed capability errors (RI-N2, SDLC-D-035).
+ *
+ * MACP must fail closed when a required capability (executor, reviewer,
+ * command, CI provider, human authority) is absent. These typed codes mirror
+ * the Forge failure vocabulary (FORGE_NO_*) so both packages speak the same
+ * language: an unimplemented capability is a failure, never a stub success.
+ */
+
+/** Closed set of typed MACP capability error codes. */
+export const MACP_ERROR_CODES = [
+  'MACP_NOT_IMPLEMENTED',
+  'MACP_NO_COMMAND',
+  'MACP_NO_REVIEWER',
+  'MACP_NO_CI_PIPELINE',
+  'MACP_NO_PROVIDER',
+  'MACP_AUTHORITY_REQUIRED',
+] as const;
+
+/** Raised when a required capability is missing and execution must fail closed. */
+export class MACPCapabilityError extends Error {
+  /** Typed error code from the closed MACP_ERROR_CODES set. */
+  readonly code: MacpErrorCode;
+  /** The missing capability, e.g. `ci-provider`, `task-persistence`, `command`. */
+  readonly capability: string;
+
+  constructor(code: MacpErrorCode, capability: string, message: string) {
+    super(message);
+    this.name = 'MACPCapabilityError';
+    this.code = code;
+    this.capability = capability;
+  }
+}

packages/macp/src/gate-runner.spec.ts

@@ -0,0 +1,429 @@
+import fs from 'node:fs';
+import os from 'node:os';
+import path from 'node:path';
+import { afterEach, beforeEach, describe, expect, it } from 'vitest';
+
+import { countAIFindings, normalizeGate, runGate, runGates } from './gate-runner.js';
+
+function makeTmpDir(): string {
+  return fs.mkdtempSync(path.join(os.tmpdir(), 'macp-gate-'));
+}
+
+describe('normalizeGate', () => {
+  it('normalizes a string to mechanical gate', () => {
+    expect(normalizeGate('echo test')).toEqual({
+      command: 'echo test',
+      type: 'mechanical',
+      fail_on: 'blocker',
+    });
+  });
+
+  it('normalizes an object gate with defaults', () => {
+    expect(normalizeGate({ command: 'lint' })).toEqual({
+      command: 'lint',
+      type: 'mechanical',
+      fail_on: 'blocker',
+    });
+  });
+
+  it('preserves explicit type and fail_on', () => {
+    expect(normalizeGate({ command: 'review', type: 'ai-review', fail_on: 'any' })).toEqual({
+      command: 'review',
+      type: 'ai-review',
+      fail_on: 'any',
+    });
+  });
+
+  it('handles non-string/non-object input', () => {
+    expect(normalizeGate(42)).toEqual({ command: '', type: 'mechanical', fail_on: 'blocker' });
+    expect(normalizeGate(null)).toEqual({ command: '', type: 'mechanical', fail_on: 'blocker' });
+  });
+});
+
+describe('countAIFindings', () => {
+  it('returns zeros for non-object', () => {
+    expect(countAIFindings(null)).toEqual({ blockers: 0, total: 0 });
+    expect(countAIFindings('string')).toEqual({ blockers: 0, total: 0 });
+    expect(countAIFindings([])).toEqual({ blockers: 0, total: 0 });
+  });
+
+  it('counts from stats block', () => {
+    const output = { stats: { blockers: 2, should_fix: 3, suggestions: 1 } };
+    expect(countAIFindings(output)).toEqual({ blockers: 2, total: 6 });
+  });
+
+  it('counts from findings array when stats has no blockers', () => {
+    const output = {
+      stats: { blockers: 0 },
+      findings: [{ severity: 'blocker' }, { severity: 'warning' }, { severity: 'blocker' }],
+    };
+    expect(countAIFindings(output)).toEqual({ blockers: 2, total: 3 });
+  });
+
+  it('uses stats blockers over findings array when stats has blockers', () => {
+    const output = {
+      stats: { blockers: 5 },
+      findings: [{ severity: 'blocker' }, { severity: 'warning' }],
+    };
+    // stats.blockers = 5, total from stats = 5+0+0 = 5, findings not used for total since stats total is non-zero
+    expect(countAIFindings(output)).toEqual({ blockers: 5, total: 5 });
+  });
+
+  it('counts findings length as total when stats has zero total', () => {
+    const output = {
+      findings: [{ severity: 'warning' }, { severity: 'info' }],
+    };
+    expect(countAIFindings(output)).toEqual({ blockers: 0, total: 2 });
+  });
+});
+
+describe('runGate', () => {
+  let tmp: string;
+  let logPath: string;
+
+  beforeEach(() => {
+    tmp = makeTmpDir();
+    logPath = path.join(tmp, 'gate.log');
+  });
+
+  afterEach(() => {
+    fs.rmSync(tmp, { recursive: true, force: true });
+  });
+
+  it('passes mechanical gate on exit 0', () => {
+    const result = runGate('echo hello', tmp, logPath, 30);
+    expect(result.passed).toBe(true);
+    expect(result.exit_code).toBe(0);
+    expect(result.type).toBe('mechanical');
+    expect(result.output).toContain('hello');
+  });
+
+  it('fails mechanical gate on non-zero exit', () => {
+    const result = runGate('exit 1', tmp, logPath, 30);
+    expect(result.passed).toBe(false);
+    expect(result.exit_code).toBe(1);
+  });
+
+  it('ci-pipeline fails closed without a CI provider (no placeholder pass)', () => {
+    const result = runGate({ command: 'anything', type: 'ci-pipeline' }, tmp, logPath, 30);
+    expect(result.passed).toBe(false);
+    expect(result.status).toBe('capability_failure');
+    expect(result.capability_code).toBe('MACP_NO_CI_PIPELINE');
+    expect(result.type).toBe('ci-pipeline');
+    expect(result.output).not.toBe('CI pipeline gate placeholder');
+  });
+
+  it('empty command is a typed capability failure, never a pass', () => {
+    const result = runGate({ command: '' }, tmp, logPath, 30);
+    expect(result.passed).toBe(false);
+    expect(result.status).toBe('capability_failure');
+    expect(result.capability_code).toBe('MACP_NO_COMMAND');
+  });
+
+  it('ai-review gate parses JSON output', () => {
+    const json = JSON.stringify({ stats: { blockers: 0, should_fix: 1 } });
+    const result = runGate({ command: `echo '${json}'`, type: 'ai-review' }, tmp, logPath, 30);
+    expect(result.passed).toBe(true);
+    expect(result.blockers).toBe(0);
+    expect(result.findings).toBe(1);
+  });
+
+  it('ai-review gate fails on blockers', () => {
+    const json = JSON.stringify({ stats: { blockers: 2 } });
+    const result = runGate({ command: `echo '${json}'`, type: 'ai-review' }, tmp, logPath, 30);
+    expect(result.passed).toBe(false);
+    expect(result.blockers).toBe(2);
+  });
+
+  it('ai-review gate with fail_on=any fails on any findings', () => {
+    const json = JSON.stringify({ stats: { blockers: 0, should_fix: 1 } });
+    const result = runGate(
+      { command: `echo '${json}'`, type: 'ai-review', fail_on: 'any' },
+      tmp,
+      logPath,
+      30,
+    );
+    expect(result.passed).toBe(false);
+    expect(result.fail_on).toBe('any');
+  });
+
+  it('ai-review gate fails on invalid JSON output', () => {
+    const result = runGate({ command: 'echo "not json"', type: 'ai-review' }, tmp, logPath, 30);
+    expect(result.passed).toBe(false);
+    expect(result.parse_error).toBeDefined();
+  });
+
+  it('writes to log file', () => {
+    runGate('echo logged', tmp, logPath, 30);
+    const log = fs.readFileSync(logPath, 'utf-8');
+    expect(log).toContain('COMMAND: echo logged');
+    expect(log).toContain('logged');
+    expect(log).toContain('EXIT:');
+  });
+});
+
+describe('runGates', () => {
+  let tmp: string;
+  let logPath: string;
+  let eventsPath: string;
+
+  beforeEach(() => {
+    tmp = makeTmpDir();
+    logPath = path.join(tmp, 'gates.log');
+    eventsPath = path.join(tmp, 'events.ndjson');
+  });
+
+  afterEach(() => {
+    fs.rmSync(tmp, { recursive: true, force: true });
+  });
+
+  it('runs multiple gates and returns results', () => {
+    const { allPassed, gateResults } = runGates(
+      ['echo one', 'echo two'],
+      tmp,
+      logPath,
+      30,
+      eventsPath,
+      'task-1',
+    );
+    expect(allPassed).toBe(true);
+    expect(gateResults).toHaveLength(2);
+  });
+
+  it('reports failure when any gate fails', () => {
+    const { allPassed, gateResults } = runGates(
+      ['echo ok', 'exit 1'],
+      tmp,
+      logPath,
+      30,
+      eventsPath,
+      'task-2',
+    );
+    expect(allPassed).toBe(false);
+    expect(gateResults[0]!.passed).toBe(true);
+    expect(gateResults[1]!.passed).toBe(false);
+  });
+
+  it('emits events for each gate', () => {
+    runGates(['echo test'], tmp, logPath, 30, eventsPath, 'task-3');
+    const events = fs
+      .readFileSync(eventsPath, 'utf-8')
+      .trim()
+      .split('\n')
+      .map((l) => JSON.parse(l));
+    expect(events).toHaveLength(2); // started + passed
+    expect(events[0].event_type).toBe('rail.check.started');
+    expect(events[1].event_type).toBe('rail.check.passed');
+  });
+
+  it('does not silently skip gates with empty command — they become capability failures', () => {
+    const { gateResults, allPassed, state } = runGates(
+      [{ command: '', type: 'mechanical' }, 'echo real'],
+      tmp,
+      logPath,
+      30,
+      eventsPath,
+      'task-4',
+    );
+    expect(gateResults).toHaveLength(2);
+    expect(gateResults[0]!.status).toBe('capability_failure');
+    expect(gateResults[1]!.status).toBe('passed');
+    expect(allPassed).toBe(false);
+    expect(state).toBe('capability_failure');
+  });
+
+  it('does not skip ci-pipeline even with empty command — typed capability failure', () => {
+    const { gateResults, allPassed, state } = runGates(
+      [{ command: '', type: 'ci-pipeline' }],
+      tmp,
+      logPath,
+      30,
+      eventsPath,
+      'task-5',
+    );
+    expect(gateResults).toHaveLength(1);
+    expect(gateResults[0]!.passed).toBe(false);
+    expect(gateResults[0]!.status).toBe('capability_failure');
+    expect(allPassed).toBe(false);
+    expect(state).toBe('capability_failure');
+  });
+
+  it('emits failed event with correct message', () => {
+    runGates(['exit 42'], tmp, logPath, 30, eventsPath, 'task-6');
+    const events = fs
+      .readFileSync(eventsPath, 'utf-8')
+      .trim()
+      .split('\n')
+      .map((l) => JSON.parse(l));
+    const failEvent = events.find(
+      (e: Record<string, unknown>) => e.event_type === 'rail.check.failed',
+    );
+    expect(failEvent).toBeDefined();
+    expect(failEvent.message).toContain('Gate failed (');
+  });
+});
+
+/**
+ * RI-N2 / SDLC-D-035 fail-closed controls for the MACP gate runner.
+ *
+ * Invariant under test: `passed: true` occurs ONLY when a gate really executed
+ * and really exited green (`status === 'passed'`). Absent capabilities,
+ * manual sign-offs, and simulated runs are typed distinctly and can never
+ * make the aggregate `passed`.
+ */
+describe('gate-runner fail-closed (RI-N2)', () => {
+  let tmpDir: string;
+  let logPath: string;
+  let eventsPath: string;
+
+  beforeEach(() => {
+    tmpDir = makeTmpDir();
+    logPath = path.join(tmpDir, 'gate.log');
+    eventsPath = path.join(tmpDir, 'events.ndjson');
+  });
+
+  afterEach(() => {
+    fs.rmSync(tmpDir, { recursive: true, force: true });
+  });
+
+  function run(gates: unknown[], options?: { simulate?: boolean }) {
+    return runGates(gates, tmpDir, logPath, 10, eventsPath, 'spec-task', options);
+  }
+
+  // ─── positive controls ───────────────────────────────────────────────────
+
+  it('a really-executed green command gate still passes', () => {
+    const result = run([{ command: 'exit 0', type: 'mechanical' }]);
+    expect(result.gateResults[0]!.status).toBe('passed');
+    expect(result.gateResults[0]!.passed).toBe(true);
+    expect(result.allPassed).toBe(true);
+    expect(result.state).toBe('passed');
+  });
+
+  it('explicit simulate completes and types every result simulated', () => {
+    const result = run([{ command: 'exit 0', type: 'mechanical' }, 'echo hello'], {
+      simulate: true,
+    });
+    expect(result.gateResults).toHaveLength(2);
+    for (const gate of result.gateResults) {
+      expect(gate.status).toBe('simulated');
+      expect(gate.passed).toBe(false);
+    }
+    expect(result.state).toBe('simulated');
+  });
+
+  it('a really-executed red command gate fails with typed status failed', () => {
+    const result = run([{ command: 'exit 3', type: 'mechanical' }]);
+    expect(result.gateResults[0]!.status).toBe('failed');
+    expect(result.gateResults[0]!.passed).toBe(false);
+    expect(result.allPassed).toBe(false);
+    expect(result.state).toBe('failed');
+  });
+
+  // ─── negative controls — each asserts typed status AND aggregate not passed ──
+
+  it('an empty-command gate is a capability_failure, not skipped and not passed', () => {
+    const result = run([{ command: '', type: 'mechanical' }]);
+    // runGates must not silently skip it — it produces a typed result
+    expect(result.gateResults).toHaveLength(1);
+    const gate = result.gateResults[0]!;
+    expect(gate.status).toBe('capability_failure');
+    expect(gate.capability_code).toBe('MACP_NO_COMMAND');
+    expect(gate.passed).toBe(false);
+    // aggregate is not passed
+    expect(result.allPassed).toBe(false);
+    expect(result.state).toBe('capability_failure');
+    expect(result.state).not.toBe('passed');
+  });
+
+  it('a commandless ai-review gate is a typed MACP_NO_REVIEWER capability_failure', () => {
+    const result = run([{ command: '', type: 'ai-review' }]);
+    expect(result.gateResults[0]!.status).toBe('capability_failure');
+    expect(result.gateResults[0]!.capability_code).toBe('MACP_NO_REVIEWER');
+    expect(result.allPassed).toBe(false);
+    expect(result.state).not.toBe('passed');
+  });
+
+  it('a ci-pipeline gate without a provider implementation is a capability_failure, never a placeholder pass', () => {
+    const result = run([{ command: '', type: 'ci-pipeline' }]);
+    const gate = result.gateResults[0]!;
+    expect(gate.status).toBe('capability_failure');
+    expect(gate.capability_code).toBe('MACP_NO_CI_PIPELINE');
+    expect(gate.passed).toBe(false);
+    // the old false-success placeholder must be gone
+    expect(gate.output).not.toBe('CI pipeline gate placeholder');
+    expect(result.allPassed).toBe(false);
+    expect(result.state).not.toBe('passed');
+  });
+
+  it('a ci-pipeline gate fails closed even alongside an otherwise green run', () => {
+    const result = run(['exit 0', { type: 'ci-pipeline', command: 'fake-ci' }]);
+    expect(result.gateResults[1]!.status).toBe('capability_failure');
+    expect(result.gateResults[0]!.status).toBe('passed');
+    expect(result.allPassed).toBe(false);
+    expect(result.state).toBe('capability_failure');
+  });
+
+  it('a manual gate with no automation enters typed waiting — neither pass nor fail', () => {
+    const result = run([{ type: 'manual' }]);
+    const gate = result.gateResults[0]!;
+    expect(gate.status).toBe('waiting');
+    expect(gate.passed).toBe(false);
+    expect(gate.exit_code).toBe(0);
+    // aggregate is not passed while any gate is waiting
+    expect(result.allPassed).toBe(false);
+    expect(result.state).toBe('waiting');
+    expect(result.state).not.toBe('passed');
+  });
+
+  it('a simulated result can never make the aggregate passed', () => {
+    const result = run(['exit 0', 'exit 0'], { simulate: true });
+    expect(result.gateResults.every((g) => g.status === 'simulated')).toBe(true);
+    expect(result.allPassed).toBe(false);
+    expect(result.state).toBe('simulated');
+    expect(result.state).not.toBe('passed');
+  });
+
+  it('waiting dominates an otherwise green aggregate', () => {
+    const result = run(['exit 0', { type: 'manual' }]);
+    expect(result.allPassed).toBe(false);
+    expect(result.state).toBe('waiting');
+  });
+});
+
+describe('runGate fail-closed (RI-N2)', () => {
+  let tmpDir: string;
+  let logPath: string;
+
+  beforeEach(() => {
+    tmpDir = makeTmpDir();
+    logPath = path.join(tmpDir, 'gate.log');
+  });
+
+  afterEach(() => {
+    fs.rmSync(tmpDir, { recursive: true, force: true });
+  });
+
+  it('simulate: true returns a typed simulated result without executing', () => {
+    const result = runGate('this-command-does-not-exist-xyz', tmpDir, logPath, 10, {
+      simulate: true,
+    });
+    expect(result.status).toBe('simulated');
+    expect(result.passed).toBe(false);
+    expect(result.exit_code).toBe(0);
+  });
+
+  it('normal mode executes for real and types a green gate passed', () => {
+    const result = runGate('echo ok', tmpDir, logPath, 10);
+    expect(result.status).toBe('passed');
+    expect(result.passed).toBe(true);
+    expect(result.output).toContain('ok');
+  });
+
+  it('a bare string gate normalizes to mechanical and executes', () => {
+    const result = runGate('exit 7', tmpDir, logPath, 10);
+    expect(result.type).toBe('mechanical');
+    expect(result.status).toBe('failed');
+    expect(result.passed).toBe(false);
+  });
+});

packages/macp/src/gate-runner.ts

@@ -4,7 +4,20 @@ import { dirname } from 'node:path';
 
 import { emitEvent } from './event-emitter.js';
 import { nowISO } from './event-emitter.js';
-import type { GateResult } from './types.js';
+import type { GateResult, GateStatus, RunGatesResult } from './types.js';
+
+/** Typed reason stamped on every simulated gate result. */
+export const SIMULATED_GATE_REASON =
+  'simulated execution (explicit simulate opt-in): gate was not evaluated by a real implementation';
+
+/** Options for gate execution (RI-N2 fail-closed / explicit simulation). */
+export interface RunGateOptions {
+  /**
+   * Explicit caller opt-in to simulation. Simulated gates are NOT executed;
+   * every result is typed `simulated` and never satisfies anything.
+   */
+  simulate?: boolean;
+}
 
 export interface NormalizedGate {
   command: string;
@@ -103,36 +116,91 @@ export function countAIFindings(parsedOutput: unknown): { blockers: number; tota
   return { blockers, total };
 }
 
+function simulatedResult(gateEntry: NormalizedGate): GateResult {
+  return {
+    command: gateEntry.command,
+    exit_code: 0,
+    type: gateEntry.type,
+    output: SIMULATED_GATE_REASON,
+    timed_out: false,
+    passed: false,
+    status: 'simulated',
+    reason: SIMULATED_GATE_REASON,
+  };
+}
+
+function capabilityFailureResult(
+  gateEntry: NormalizedGate,
+  code: GateResult['capability_code'],
+  reason: string,
+): GateResult {
+  return {
+    command: gateEntry.command,
+    exit_code: 1,
+    type: gateEntry.type,
+    output: '',
+    timed_out: false,
+    passed: false,
+    status: 'capability_failure',
+    capability_code: code,
+    reason,
+  };
+}
+
+function waitingResult(gateEntry: NormalizedGate, reason: string): GateResult {
+  return {
+    command: gateEntry.command,
+    exit_code: 0,
+    type: gateEntry.type,
+    output: '',
+    timed_out: false,
+    passed: false,
+    status: 'waiting',
+    capability_code: 'MACP_AUTHORITY_REQUIRED',
+    reason,
+  };
+}
+
 export function runGate(
   gate: unknown,
   cwd: string,
   logPath: string,
   timeoutSec: number,
+  options: RunGateOptions = {},
 ): GateResult {
   const gateEntry = normalizeGate(gate);
   const gateType = gateEntry.type;
   const command = gateEntry.command;
 
+  // Explicit simulation only: never executes, typed simulated, never satisfying.
+  if (options.simulate) {
+    return simulatedResult(gateEntry);
+  }
+
+  // Fail closed: no CI provider implementation exists in @mosaicstack/macp,
+  // so a ci-pipeline gate is an absent capability — never a placeholder pass.
   if (gateType === 'ci-pipeline') {
-    return {
-      command,
-      exit_code: 0,
-      type: gateType,
-      output: 'CI pipeline gate placeholder',
-      timed_out: false,
-      passed: true,
-    };
+    return capabilityFailureResult(
+      gateEntry,
+      'MACP_NO_CI_PIPELINE',
+      `ci-pipeline gate '${gateEntry.command || gateType}' has no CI provider implementation wired — refusing placeholder pass`,
+    );
   }
 
   if (!command) {
-    return {
-      command: '',
-      exit_code: 0,
-      type: gateType,
-      output: '',
-      timed_out: false,
-      passed: true,
-    };
+    // A manual gate with no automation waits for human sign-off: not pass, not fail.
+    if (gateType === 'manual') {
+      return waitingResult(
+        gateEntry,
+        `manual gate has no automation — waiting for human sign-off (type: ${gateType})`,
+      );
+    }
+    // Any other commandless gate is an absent capability — never a vacuous pass.
+    return capabilityFailureResult(
+      gateEntry,
+      gateType === 'ai-review' ? 'MACP_NO_REVIEWER' : 'MACP_NO_COMMAND',
+      `gate of type '${gateType}' has no command to execute — refusing empty-command pass`,
+    );
   }
 
   const { exitCode, output, timedOut } = runShell(command, cwd, logPath, timeoutSec);
@@ -143,10 +211,12 @@ export function runGate(
     output,
     timed_out: timedOut,
     passed: false,
+    status: 'failed',
   };
 
   if (gateType !== 'ai-review') {
     result.passed = exitCode === 0;
+    result.status = result.passed ? 'passed' : 'failed';
     return result;
   }
 
@@ -170,6 +240,7 @@ export function runGate(
   } else {
     result.passed = exitCode === 0 && blockers === 0 && !timedOut && parseError === undefined;
   }
+  result.status = result.passed ? 'passed' : 'failed';
 
   result.fail_on = failOn;
   result.blockers = blockers;
@@ -191,16 +262,19 @@ export function runGates(
   timeoutSec: number,
   eventsPath: string,
   taskId: string,
-): { allPassed: boolean; gateResults: GateResult[] } {
-  let allPassed = true;
+  options: RunGateOptions = {},
+): RunGatesResult {
   const gateResults: GateResult[] = [];
+  let hasCapabilityFailure = false;
+  let hasSimulated = false;
+  let hasFailed = false;
+  let hasWaiting = false;
 
   for (const gate of gates) {
     const gateEntry = normalizeGate(gate);
     const gateCmd = gateEntry.command;
-    if (!gateCmd && gateEntry.type !== 'ci-pipeline') continue;
-
     const label = gateCmd || gateEntry.type;
+    // NOTE: no silent skip — every gate produces a typed result (RI-N2).
     emitEvent(
       eventsPath,
       'rail.check.started',
@@ -209,10 +283,10 @@ export function runGates(
       'quality-gate',
       `Running gate: ${label}`,
     );
-    const result = runGate(gate, cwd, logPath, timeoutSec);
+    const result = runGate(gate, cwd, logPath, timeoutSec, options);
     gateResults.push(result);
 
-    if (result.passed) {
+    if (result.status === 'passed') {
       emitEvent(
         eventsPath,
         'rail.check.passed',
@@ -224,7 +298,46 @@ export function runGates(
       continue;
     }
 
-    allPassed = false;
+    if (result.status === 'waiting') {
+      hasWaiting = true;
+      emitEvent(
+        eventsPath,
+        'rail.check.waiting',
+        taskId,
+        'gated',
+        'quality-gate',
+        `Gate waiting: ${label} — ${result.reason ?? 'manual gate awaits sign-off'}`,
+      );
+      continue;
+    }
+
+    if (result.status === 'simulated') {
+      hasSimulated = true;
+      emitEvent(
+        eventsPath,
+        'rail.check.simulated',
+        taskId,
+        'gated',
+        'quality-gate',
+        `Gate simulated (non-satisfying): ${label}`,
+      );
+      continue;
+    }
+
+    if (result.status === 'capability_failure') {
+      hasCapabilityFailure = true;
+      emitEvent(
+        eventsPath,
+        'rail.check.failed',
+        taskId,
+        'gated',
+        'quality-gate',
+        `Gate capability failure (${result.capability_code ?? 'MACP_NO_PROVIDER'}): ${label} — ${result.reason ?? 'required capability is absent'}`,
+      );
+      continue;
+    }
+
+    hasFailed = true;
     let message: string;
     if (result.timed_out) {
       message = `Gate timed out after ${timeoutSec}s: ${label}`;
@@ -236,5 +349,15 @@ export function runGates(
     emitEvent(eventsPath, 'rail.check.failed', taskId, 'gated', 'quality-gate', message);
   }
 
-  return { allPassed, gateResults };
+  const state: GateStatus = hasCapabilityFailure
+    ? 'capability_failure'
+    : hasSimulated
+      ? 'simulated'
+      : hasFailed
+        ? 'failed'
+        : hasWaiting
+          ? 'waiting'
+          : 'passed';
+
+  return { allPassed: state === 'passed', gateResults, state };
 }

packages/macp/src/index.ts

@@ -6,11 +6,13 @@ export type {
   DependsOnPolicy,
   GateType,
   GateFailOn,
+  GateStatus,
   GateEntry,
   Task,
   EventType,
   MACPEvent,
   GateResult,
+  RunGatesResult,
   TaskResult,
   ProviderMeta,
   ProviderRegistry,
@@ -18,6 +20,11 @@ export type {
 
 export { CredentialError } from './types.js';
 
+// Typed fail-closed capability errors (RI-N2, SDLC-D-035)
+export { MACP_ERROR_CODES, MACPCapabilityError } from './errors.js';
+
+export type { MacpErrorCode } from './errors.js';
+
 // Credential resolver
 export {
   DEFAULT_CREDENTIALS_DIR,
@@ -35,9 +42,16 @@ export {
 export type { ResolveCredentialsOptions } from './credential-resolver.js';
 
 // Gate runner
-export { normalizeGate, runShell, countAIFindings, runGate, runGates } from './gate-runner.js';
+export {
+  normalizeGate,
+  runShell,
+  countAIFindings,
+  runGate,
+  runGates,
+  SIMULATED_GATE_REASON,
+} from './gate-runner.js';
 
-export type { NormalizedGate } from './gate-runner.js';
+export type { NormalizedGate, RunGateOptions } from './gate-runner.js';
 
 // Risk-floor (agent reflection loop — diff review classifier)
 export { evaluateRiskFloor, DEFAULT_RISK_THRESHOLD } from './risk-floor.js';

packages/macp/src/types.ts

@@ -1,3 +1,5 @@
+import type { MacpErrorCode } from './errors.js';
+
 /** Task status values. */
 export type TaskStatus = 'pending' | 'running' | 'gated' | 'completed' | 'failed' | 'escalated';
 
@@ -17,7 +19,17 @@ export type DispatchMode = 'yolo' | 'acp' | 'exec';
 export type DependsOnPolicy = 'all' | 'any' | 'all_terminal';
 
 /** Quality gate type. */
-export type GateType = 'mechanical' | 'ai-review' | 'ci-pipeline';
+export type GateType = 'mechanical' | 'ai-review' | 'ci-pipeline' | 'manual';
+
+/**
+ * Typed execution state of a gate — closed set (RI-N2, SDLC-D-035).
+ *
+ * Only `passed` means "really executed and green". `simulated` is produced
+ * exclusively under an explicit simulate opt-in and never satisfies anything.
+ * `capability_failure` means a required executor/provider/command was absent.
+ * `waiting` means a manual gate awaits human sign-off (neither pass nor fail).
+ */
+export type GateStatus = 'passed' | 'failed' | 'simulated' | 'waiting' | 'capability_failure';
 
 /** Gate fail_on mode. */
 export type GateFailOn = 'blocker' | 'any';
@@ -67,7 +79,9 @@ export type EventType =
   | 'task.retry.scheduled'
   | 'rail.check.started'
   | 'rail.check.passed'
-  | 'rail.check.failed';
+  | 'rail.check.failed'
+  | 'rail.check.waiting'
+  | 'rail.check.simulated';
 
 /** Structured event record. */
 export interface MACPEvent {
@@ -88,7 +102,14 @@ export interface GateResult {
   type: string;
   output: string;
   timed_out: boolean;
+  /** Back-compat boolean view — true ONLY when `status === 'passed'`. */
   passed: boolean;
+  /** Typed discriminator — the authoritative gate outcome (RI-N2). */
+  status: GateStatus;
+  /** Typed capability error code, set when `status === 'capability_failure'`. */
+  capability_code?: MacpErrorCode;
+  /** Why a non-executed state (simulated/waiting/capability_failure) was reached. */
+  reason?: string;
   fail_on?: string;
   blockers?: number;
   findings?: number;
@@ -96,6 +117,22 @@ export interface GateResult {
   parse_error?: string;
 }
 
+/**
+ * Aggregate outcome of `runGates` (RI-N2).
+ *
+ * `state` is the typed aggregate: it is `passed` only when every gate really
+ * executed green. A `simulated` result makes the aggregate `simulated` (never
+ * `passed`); a `waiting` manual gate keeps the aggregate `waiting`; a missing
+ * capability makes it `capability_failure`. `allPassed` is exactly
+ * `state === 'passed'`, so a simulated or waiting result can never satisfy a
+ * dependency, acceptance criterion, gate, merge, or release check.
+ */
+export interface RunGatesResult {
+  allPassed: boolean;
+  gateResults: GateResult[];
+  state: GateStatus;
+}
+
 /** Result from a completed task. */
 export interface TaskResult {
   task_id: string;

packages/mosaic/framework/tools/quality/test-enumeration-exclusions.txt

@@ -39,3 +39,20 @@ packages/mosaic/framework/tools/tmux/test-send-message-verdict.sh | requires rea
 # recorded judgement. These lines ARE that judgement, signed.)
 packages/mosaic/framework/tools/orchestrator/smoke-test.sh | behavior smoke checks for coord continue/run workflows, run manually by orchestrator seats; unmeasured in CI; #1017 burndown
 packages/mosaic/framework/tools/wake/validate-973/microtest-wake-assert.sh | #973 instrument self-test, run as a precondition of the validate-973 evidence procedure rather than as a standing CI suite; #1017 burndown candidate
+
+# --- tools/fleet: precondition is unsatisfiable in the CI image (#1271) ---
+# Signed by fred (sb-it-1-dt, 2026-08-16) at origin/next 476db12.
+# This suite asserts the launcher's behaviour when `mosaic` and `pi` are MISSING.
+# It shims fakes into $FAKE_BIN, but the constructed PANE_PATH always ends in the
+# real system path, so on a host that installs those binaries the missing-binary
+# cases cannot be measured at all. The suite's own guard (line 103) says so and
+# fails rather than reporting a pass it cannot back. That guard is correct.
+# The error was wiring the suite into CI: #1017 (c56483eb) enumerated it and
+# dropped this exclusion, and the CI image provides `pi` in the system path, so
+# it has failed on every pipeline since. Measured 2026-08-16 across pipelines
+# 2444 (#1256), 2438 (#1240) and 2441 (#1017-quality): exactly one FAIL line in
+# each full log, identical, this assertion; control `zzz-not-present-zzz` -> 0.
+# Burn-down and the full measurement are tracked in #1271; unwired by PR #1270.
+# Because test:framework-shell is one && chain and this sat at position 44 of 48,
+# the four suites after it had not run at all since the merge.
+packages/mosaic/framework/tools/fleet/test-start-agent-session.sh | precondition unsatisfiable in the CI image: asserts missing-binary behaviour, but PANE_PATH always ends in the system path and the image provides `pi` there; guard at line 103 fails by design rather than passing unmeasured. Burn down by controlling the tail of PANE_PATH inside the test. NOT by removing `pi` from the image: the CI image installs @earendil-works/[email protected] deliberately (measured in pipeline 2444's test-step log), and other suites depend on that pin. Burn-down tracked in #1271

packages/mosaic/framework/tools/tmux/send-message.sh

@@ -97,34 +97,13 @@ printf '%s' "$MSG" | "${tmux_cmd[@]}" load-buffer -b "$BUF" -
 #   would otherwise accumulate forever.
 sleep 0.5
 
-# 2) Submit, then POSITIVELY confirm submission by DRAFT TRANSITION, not by prompt
-#    glyph. The historical bug was treating ABSENCE of a draft as delivery; the
-#    2026-08 fix over-corrected to glyph inference (grep '❯|^>|│ >'), which locates
-#    only Claude Code's box and false-NEGATIVES every glyphless REPL (pi renders a
-#    U+2500 rule, no glyph) — a delivered message reported "UNDELIVERED", driving a
-#    retry that duplicates it. Runtime-agnostic evidence: our message tail sits on
-#    the INPUT line (located by the cursor row, not a glyph) BEFORE Enter, and has
-#    LEFT it AFTER — that transition is positive proof of submission and needs no
-#    glyph. Absence alone still never means delivered: if we never saw our draft on
-#    the input line we stay UNCONFIRMED (wrong/dead pane), and a draft that never
-#    leaves the input line stays a DRAFT (exit 2), preserving both historical guards.
-_cursor_line() { # echo the pane's current input (cursor) line, glyph-free
-  local cy line
-  cy=$("${tmux_cmd[@]}" display-message -p -t "$EFFECTIVE_TARGET" -F '#{cursor_y}' 2>/dev/null) || return 1
-  [ -n "$cy" ] || return 1
-  "${tmux_cmd[@]}" capture-pane -t "$EFFECTIVE_TARGET" -p 2>/dev/null | sed -n "$((cy + 1))p"
-}
-_draft_on_input() { # true iff our message tail is sitting on the input line now
-  [ -n "$snippet" ] || return 1
-  printf '%s' "$(_cursor_line)" | grep -qF "$snippet"
-}
-
-# Baseline: after the paste, our draft must be on the input line. This is positive
-# proof we are on the right pane and the paste landed — the anchor the transition
-# check measures against.
-saw_draft=0
-_draft_on_input && saw_draft=1
-
+# 2) Submit, then POSITIVELY confirm submission; flush with another Enter if it is
+#    still a draft. Success requires positive evidence — the queued banner, OR the
+#    REPL input box located AND clear of our message tail. The historical bug was
+#    treating ABSENCE of a draft as delivery: if the prompt glyph was never matched
+#    (wrong pane / prompt-glyph drift), an unsubmitted message read as "delivered"
+#    and worker->lead relays stalled silently. We now default to UNCONFIRMED and only
+#    upgrade to delivered on positive evidence; anything we cannot confirm fails loud.
 status="unconfirmed"
 for attempt in $(seq 1 $((RETRIES + 1))); do
   "${tmux_cmd[@]}" send-keys -t "$EFFECTIVE_TARGET" Enter
@@ -134,26 +113,20 @@ for attempt in $(seq 1 $((RETRIES + 1))); do
   if printf '%s' "$pane" | grep -qF "$QUEUED_RE"; then
     status="queued"; break
   fi
-  # POSITIVE draft evidence from a located prompt box, when one exists. This is the
-  # cursor-row check's blind spot: a pane in COOKED mode (a plain shell whose
-  # foreground process never reads stdin) echoes our paste via the kernel line
-  # discipline and moves the cursor off it on Enter, which is indistinguishable from
-  # a real submit by cursor row alone. If a prompt box IS locatable and still carries
-  # our tail, that is affirmative proof the message was not consumed. Absence of a
-  # glyph is still never used for anything — that inference is the original E7 bug.
+  # Locate the REPL input box (prompt glyph). If we cannot see it, we have NO
+  # evidence of submission state — stay UNCONFIRMED and retry; never infer delivery.
   promptline=$(printf '%s' "$pane" | grep -E '❯|^>|│ >' | tail -1)
-  if [ -n "$promptline" ] && [ -n "$snippet" ] && printf '%s' "$promptline" | grep -qF "$snippet"; then
+  if [ -z "$promptline" ]; then
+    status="unconfirmed"; continue
+  fi
+  # Input box located AND still carrying our tail => unsubmitted draft. Flush + retry.
+  # (Submitted messages scroll up into history; a draft stays on the ❯ line.)
+  if [ -n "$snippet" ] && printf '%s' "$promptline" | grep -qF "$snippet"; then
     status="draft"; continue
   fi
-  if [ "$saw_draft" = 1 ]; then
-    if _draft_on_input; then
-      status="draft"; continue      # still on the input line => not submitted; flush + retry
-    fi
-    status="delivered"; break        # left the input line => positively submitted
-  fi
-  # No confirmed baseline yet: try to (re)acquire it; never infer delivery from absence.
-  if _draft_on_input; then saw_draft=1; status="draft"; continue; fi
-  status="unconfirmed"; continue
+  # Input box located AND clear of our tail => positively submitted. This is the
+  # only path to success besides the queued banner.
+  status="delivered"; break
 done
 
 [ "$VERBOSE" = 1 ] && { echo "--- pane tail ($TARGET) ---"; printf '%s\n' "$pane" | tail -4; echo "---"; }

packages/mosaic/framework/tools/tmux/test-send-message-glyph-agnostic.sh

@@ -1,97 +0,0 @@
-#!/usr/bin/env bash
-# Red-first regression test for E7 (#1017 task 2): the confirm-check must bind
-# "delivered" to WHETHER THE MESSAGE WAS SUBMITTED, not to which runtime's prompt
-# glyph is present. A pi seat renders a U+2500 rule input box with no ❯/^>/│ >
-# glyph; send-message.sh:118 locates the box only by glyph, so a genuinely
-# delivered message on a glyphless REPL falsely reports exit 2 "may be UNDELIVERED",
-# and the operator's rc=2-driven retry duplicates it.
-#
-# Parameterized on $SEND: RED against the shipping blob (B and D fail), GREEN
-# against a candidate patch. No pi; no fake HOME; hermetic throwaway socket.
-#
-# Submission counting is EXACT and terminal-echo-independent: the fixture message
-# is `echo <tok> >>SINK`; each real submission appends one line. wc -l SINK ==
-# number of times the REPL actually executed the send. This does not depend on how
-# many times the marker string is painted on screen.
-set -u
-SEND="${SEND:?set SEND=/path/to/send-message.sh}"
-SOCKET="glyphagnostic-$$"
-TMP="$(mktemp -d)"
-tmux() { command tmux -L "$SOCKET" "$@"; }
-cleanup() { command tmux -L "$SOCKET" kill-server 2>/dev/null; rm -rf "$TMP"; }
-trap cleanup EXIT
-pass=0; fail=0
-ok() { printf 'ok   %s\n' "$1"; pass=$((pass+1)); }
-no() { printf 'FAIL %s -- %s\n' "$1" "$2"; fail=$((fail+1)); }
-
-mk() { tmux new-session -d -s "$1" -x 120 -y 40 -c "$TMP" "PS1='$2' exec bash --noprofile --norc -i"; sleep 0.5; }
-subs() { [ -f "$1" ] && wc -l <"$1" | tr -d ' ' || echo 0; }  # exact submission count
-
-echo "SEND=$SEND    tmux $(command tmux -V | awk '{print $2}')"
-
-# --- A (control): glyph box (❯) that submits => exit 0, exactly one submission.
-mk ctl '❯ '
-SINK="$TMP/sink.ctl"
-out=$("$SEND" -L "$SOCKET" -t ctl -m "echo x >>'$SINK'" 2>"$TMP/e.ctl"); rc=$?; sleep 0.4
-if [ "$rc" = 0 ] && [ "$(subs "$SINK")" = 1 ]; then
-  ok "control: ❯-box submits => exit 0, exactly one submission"
-else no "control: ❯-box submits => exit 0, one submission" "rc=$rc subs=$(subs "$SINK") err=[$(cat "$TMP/e.ctl")]"; fi
-
-# --- B (THE false-rc regression): glyphless U+2500 box that SUBMITS. Message lands
-#     (subs==1) yet shipping reports exit 2. Must be exit 0.
-mk sub $'──────── \n'
-SINK="$TMP/sink.sub"
-out=$("$SEND" -L "$SOCKET" -t sub -m "echo x >>'$SINK'" 2>"$TMP/e.sub"); rc=$?; sleep 0.4
-if [ "$rc" = 0 ] && [ "$(subs "$SINK")" = 1 ]; then
-  ok "glyphless: U+2500 box that submits => exit 0 (delivered, not 'UNDELIVERED')"
-else no "glyphless: U+2500 box that submits => exit 0" \
-        "rc=$rc subs=$(subs "$SINK")(delivered=$([ "$(subs "$SINK")" -ge 1 ] && echo yes||echo no)) err=[$(cat "$TMP/e.sub")]"; fi
-
-# --- D (duplicate arm): operator follows the rc=2 stderr and retries once. On the
-#     glyphless box, shipping => two submissions (the reported duplicate). The
-#     property: one logical send => exactly one submission. Same fix closes it.
-mk dup $'──────── \n'
-SINK="$TMP/sink.dup"
-tries=0
-for attempt in 1 2; do
-  tries=$((tries+1))
-  out=$("$SEND" -L "$SOCKET" -t dup -m "echo x >>'$SINK'" 2>/dev/null); rc=$?
-  sleep 0.4
-  [ "$rc" = 0 ] && break   # operator stops retrying only when told delivered
-done
-if [ "$(subs "$SINK")" = 1 ]; then
-  ok "duplicate: one logical send (rc-driven retry) => exactly one submission (tries=$tries)"
-else no "duplicate: one logical send => exactly one submission" "submissions=$(subs "$SINK") tries=$tries"; fi
-
-# --- E (faithful hung managed TUI, NOT a cooked shell): raw/no-echo, paints nothing.
-#     A cooked `sleep infinity` echoes the paste via the kernel line discipline and
-#     false-passes a cursor-row fix that is correct on real seats (measured). So: raw.
-mk_rawstuck() { tmux new-session -d -s "$1" -x 120 -y 40 -c "$TMP" \
-  "bash --noprofile --norc -c 'stty -echo -icanon min 1 time 0 2>/dev/null; exec sleep infinity'"; sleep 0.5; }
-mk_rawstuck estuck
-SINK="$TMP/sink.estuck"
-out=$("$SEND" -L "$SOCKET" -t estuck -r 1 -m "this stuck draft was never submitted" 2>/dev/null); rc=$?
-sleep 0.3
-if [ "$rc" != 0 ] && [ "$(subs "$SINK")" = 0 ]; then
-  ok "raw/no-echo stuck TUI (not submitted) => non-zero (no false delivered)"
-else no "raw stuck TUI must NOT report delivered" "rc=$rc subs=$(subs "$SINK")"; fi
-
-# --- F (busy/queued branch, your BUSY-not-runtime finding): glyphless pane rendering the
-#     queued banner, never consuming. QUEUED_RE :113 fires before the glyph grep => rc=0.
-mk_busy() { tmux new-session -d -s "$1" -x 120 -y 40 -c "$TMP" \
-  "bash --noprofile --norc -c 'printf \"Press up to edit queued messages\n\"; exec sleep infinity'"; sleep 0.5; }
-mk_busy ebusy
-SINK="$TMP/sink.ebusy"
-out=$("$SEND" -L "$SOCKET" -t ebusy -m "echo x >>'$SINK'" 2>/dev/null); rc=$?; sleep 0.3
-if [ "$rc" = 0 ]; then
-  ok "busy/queued-banner glyphless => exit 0 (queued is delivery; runtime owns custody)"
-else no "busy/queued-banner must report delivered" "rc=$rc"; fi
-
-# --- C (historical-bug guard): unresolvable target. No pane ever carried our draft
-#     => must fail, never infer delivered from absence of a glyph/snippet.
-if out=$("$SEND" -L "$SOCKET" -t "nonexistent-$$" -m "echo x >>'$TMP/sink.wrong'" 2>/dev/null); then
-  no "wrong-pane: unresolvable target must NOT report success" "expected non-zero, got 0"
-else ok "wrong-pane: unresolvable target => non-zero (no false delivered)"; fi
-
-echo "---"; echo "pass=$pass fail=$fail"
-[ "$fail" = 0 ]

packages/mosaic/framework/tools/tmux/test-send-message-verdict.sh

@@ -4,13 +4,10 @@
 #
 #   1. DELIVERED   — a REPL that renders a `❯ ` input box and submits on Enter
 #                    (text scrolls to history, box clears) => exit 0 "✓ delivered".
-#   2. DELIVERED   — a pane with NO prompt glyph that DOES submit => exit 0. A pi
-#                    seat is this fixture (U+2500 rule, no glyph). Reshaped for
-#                    #1257; see the note at the fixture for why the old exit-2
-#                    assertion was wrong.
-#   2b. UNCONFIRMED— a glyphless pane that never submits (raw/no-echo hung TUI)
-#                    => must fail loud. This carries the historical
-#                    false-positive guard that fixture 2 used to be credited with.
+#   2. UNCONFIRMED — a pane with NO locatable prompt glyph. This is the exact
+#                    historical FALSE POSITIVE: pre-patch it printed "✓ delivered"
+#                    exit 0; post-patch it MUST fail loud (exit 2, stderr
+#                    "could not confirm submission").
 #   3. DRAFT       — a `❯ `-prompt pane that never submits (message stays on the
 #                    input line) => exit 2, stderr "unsubmitted draft".
 set -uo pipefail
@@ -40,44 +37,19 @@ else
   no "delivered: ❯-prompt REPL that submits => exit 0 ✓ delivered" "rc=$rc out=[$out] err=[$(cat "$TMP/e1")]"
 fi
 
-# --- Fixture 2: NO prompt glyph, and the pane DOES submit (interactive bash).
-#     RESHAPED 2026-08-16 (#1257), deliberately. This fixture previously asserted
-#     exit 2 here and was labelled "false-positive FIXED". That assertion was wrong,
-#     and locking it in is what kept E7 alive: the pane submits, so "delivered" is
-#     the truth, and a pi seat — whose input box is a bare U+2500 rule with no glyph
-#     — IS this fixture. Reporting exit 2 for it told operators a delivered message
-#     may be undelivered, and the retry that advice invites is the duplicate.
-#
-#     The guard this fixture was reaching for is real and is NOT dropped: "never
-#     infer delivered from absence" is now enforced positively by fixture 2b below
-#     (glyphless AND not submitting => must fail) and by fixture 3 (locatable box
-#     still carrying our tail => draft). Absence alone decides nothing either way.
+# --- Fixture 2: NO prompt glyph (default bash PS1). THE regression: pre-patch this
+#     was a silent false-positive "delivered"; post-patch it must be unconfirmed→exit 2.
 tmux -L "$SOCKET" new-session -d -s noglyph -c "$TMP" \
   'PS1="sh-noglyph$ " exec bash --noprofile --norc -i'
 sleep 0.3
-out=$("$SEND" -L "$SOCKET" -t "=noglyph" -m "verdict fixture two must fail loud" 2>"$TMP/e2"); rc=$?
-if [ "$rc" -eq 0 ] && printf '%s' "$out" | grep -qF "✓ delivered"; then
-  ok "delivered: glyphless pane that submits => exit 0 (runtime-agnostic, E7 FIXED)"
-else
-  no "delivered: glyphless pane that submits => exit 0" "rc=$rc out=[$out] err=[$(cat "$TMP/e2")]"
-fi
-
-# --- Fixture 2b: NO prompt glyph AND never submits — a hung managed TUI holding the
-#     terminal in raw/no-echo, which is what a stuck agent seat actually is (measured
-#     on live pi: stty -echo -icanon). Nothing is echoed, nothing is consumed, so
-#     there is no positive evidence of submission and the tool MUST fail loud. This
-#     is the historical false-positive guard, kept as a positive test.
-tmux -L "$SOCKET" new-session -d -s rawstuck -c "$TMP" \
-  'bash --noprofile --norc -c "stty -echo -icanon min 1 time 0 2>/dev/null; exec sleep infinity"'
-sleep 0.3
-if out=$("$SEND" -L "$SOCKET" -t "=rawstuck" -r 1 -m "verdict fixture two-b never submitted" 2>"$TMP/e2b"); then
-  no "unconfirmed: glyphless hung TUI must NOT report success" "expected non-zero, got 0 (out=[$out])"
+if out=$("$SEND" -L "$SOCKET" -t "=noglyph" -m "verdict fixture two must fail loud" 2>"$TMP/e2"); then
+  no "unconfirmed: glyphless pane must NOT report success" "expected exit 2, got 0 (out=[$out])"
 else
   rc=$?
-  if [ "$rc" -ne 0 ] && grep -qF "could not confirm submission" "$TMP/e2b"; then
-    ok "unconfirmed: glyphless hung TUI (raw/no-echo) => non-zero + 'could not confirm submission'"
+  if [ "$rc" -eq 2 ] && grep -qF "could not confirm submission" "$TMP/e2"; then
+    ok "unconfirmed: glyphless pane => exit 2 + 'could not confirm submission' (false-positive FIXED)"
   else
-    no "unconfirmed: glyphless hung TUI => non-zero + stderr" "rc=$rc err=[$(cat "$TMP/e2b")]"
+    no "unconfirmed: glyphless pane => exit 2 + stderr" "rc=$rc err=[$(cat "$TMP/e2")]"
   fi
 fi
 

packages/mosaic/package.json

@@ -25,7 +25,7 @@
     "lint": "eslint src",
     "typecheck": "tsc --noEmit",
     "test": "vitest run --passWithNoTests && pnpm run test:framework-shell",
-    "test:framework-shell": "bash framework/tools/quality/scripts/check-test-enumeration.sh && bash framework/tools/quality/scripts/test-check-test-enumeration.sh && python3 src/lease-broker/daemon_deadline_unittest.py && python3 src/lease-broker/normative_fragments_unittest.py && python3 src/lease-broker/promotion_binding_unittest.py && python3 src/lease-broker/promotion_trigger_unittest.py && python3 src/lease-broker/receipt_challenge_unittest.py && python3 src/lease-broker/context_recovery_unittest.py && python3 src/lease-broker/recovery_runtime_unittest.py && python3 src/lease-broker/recovery_b1_adversarial_unittest.py && python3 src/lease-broker/receipt_observer_client_unittest.py && python3 src/lease-broker/invariant_r_unittest.py && python3 src/lease-broker/framework_skill_portability_unittest.py && python3 src/mutator-gate/runtime_tools_unittest.py && python3 src/mutator-gate/runtime_launch_guard_unittest.py && python3 src/mutator-gate/version_coupling_unittest.py && python3 framework/tools/lease-broker/check-runtime-launches.py --root ../.. && bash framework/tools/codex/test-pr-diff-context.sh && bash framework/tools/qa/test-deps-preflight.sh && bash framework/tools/git/test-pr-review-gitea-comment.sh && bash framework/tools/git/test-pr-review-repo-host-override.sh && bash framework/tools/git/test-ci-queue-wait-branch-absent.sh && bash framework/tools/git/test-ci-queue-wait-tristate.sh && bash framework/tools/git/test-ci-queue-wait-github-checks.sh && bash framework/tools/git/test-pr-merge-queue-branch.sh && bash framework/tools/git/test-pr-merge-head-pin.sh && bash framework/tools/git/test-pr-merge-message-field.sh && bash framework/tools/git/test-git-credential-mosaic.sh && bash framework/tools/git/test-gitea-token-identity.sh && bash framework/tools/woodpecker/test-terminal-green-contract.sh && bash framework/tools/_scripts/test-install-ordering-guard.sh && bash framework/tools/_scripts/test-mosaic-init-rce.sh && bash framework/tools/tmux/agent-send.test.sh && bash framework/tools/wake/test-wake-store-ack.sh && bash framework/tools/wake/test-wake-store-enqueue-race.sh && bash framework/tools/wake/test-wake-digest-hmac.sh && bash framework/tools/wake/test-wake-digest-quarantine.sh && bash framework/tools/wake/test-wake-detector.sh && bash framework/tools/wake/test-wake-fn-oracle.sh && bash framework/tools/wake/test-wake-reconcile.sh && bash framework/tools/wake/test-wake-beacon.sh && bash framework/tools/wake/test-wake-preimage.sh && bash framework/tools/wake/test-wake-install.sh && bash framework/tools/fleet/test-start-agent-session.sh && bash framework/tools/glpi/test-list-http-status.sh && bash framework/tools/orchestrator/test-board-roll.sh && bash framework/tools/woodpecker/test-ci-wait-exit-matrix.sh && bash framework/tools/_scripts/test-fleet-transport-check.sh"
+    "test:framework-shell": "bash framework/tools/quality/scripts/check-test-enumeration.sh && bash framework/tools/quality/scripts/test-check-test-enumeration.sh && python3 src/lease-broker/daemon_deadline_unittest.py && python3 src/lease-broker/normative_fragments_unittest.py && python3 src/lease-broker/promotion_binding_unittest.py && python3 src/lease-broker/promotion_trigger_unittest.py && python3 src/lease-broker/receipt_challenge_unittest.py && python3 src/lease-broker/context_recovery_unittest.py && python3 src/lease-broker/recovery_runtime_unittest.py && python3 src/lease-broker/recovery_b1_adversarial_unittest.py && python3 src/lease-broker/receipt_observer_client_unittest.py && python3 src/lease-broker/invariant_r_unittest.py && python3 src/lease-broker/framework_skill_portability_unittest.py && python3 src/mutator-gate/runtime_tools_unittest.py && python3 src/mutator-gate/runtime_launch_guard_unittest.py && python3 src/mutator-gate/version_coupling_unittest.py && python3 framework/tools/lease-broker/check-runtime-launches.py --root ../.. && bash framework/tools/codex/test-pr-diff-context.sh && bash framework/tools/qa/test-deps-preflight.sh && bash framework/tools/git/test-pr-review-gitea-comment.sh && bash framework/tools/git/test-pr-review-repo-host-override.sh && bash framework/tools/git/test-ci-queue-wait-branch-absent.sh && bash framework/tools/git/test-ci-queue-wait-tristate.sh && bash framework/tools/git/test-ci-queue-wait-github-checks.sh && bash framework/tools/git/test-pr-merge-queue-branch.sh && bash framework/tools/git/test-pr-merge-head-pin.sh && bash framework/tools/git/test-pr-merge-message-field.sh && bash framework/tools/git/test-git-credential-mosaic.sh && bash framework/tools/git/test-gitea-token-identity.sh && bash framework/tools/woodpecker/test-terminal-green-contract.sh && bash framework/tools/_scripts/test-install-ordering-guard.sh && bash framework/tools/_scripts/test-mosaic-init-rce.sh && bash framework/tools/tmux/agent-send.test.sh && bash framework/tools/wake/test-wake-store-ack.sh && bash framework/tools/wake/test-wake-store-enqueue-race.sh && bash framework/tools/wake/test-wake-digest-hmac.sh && bash framework/tools/wake/test-wake-digest-quarantine.sh && bash framework/tools/wake/test-wake-detector.sh && bash framework/tools/wake/test-wake-fn-oracle.sh && bash framework/tools/wake/test-wake-reconcile.sh && bash framework/tools/wake/test-wake-beacon.sh && bash framework/tools/wake/test-wake-preimage.sh && bash framework/tools/wake/test-wake-install.sh && bash framework/tools/glpi/test-list-http-status.sh && bash framework/tools/orchestrator/test-board-roll.sh && bash framework/tools/woodpecker/test-ci-wait-exit-matrix.sh && bash framework/tools/_scripts/test-fleet-transport-check.sh"
   },
   "dependencies": {
     "@mosaicstack/brain": "workspace:*",