harness-everything 0.47.2 → 0.47.3

package/ci/mechanism-36-verify-timeout.test.js

@@ -79,7 +79,7 @@ for (let i = 0; i < DEFAULTS.length; i += 1) {
   assert.ok(/timed out after/.test(source), `${label} labels timeout failures as timeouts`);
   assert.ok(/test_failure/.test(source), `${label} classifies confirmed failing checks`);
   assert.ok(/execution_error/.test(source), `${label} classifies spawn-level execution errors`);
-  assert.ok(/maxBuffer/.test(source), `${label} bounds captured output instead of using the 1 MiB default`);
+  assert.ok(i === 0 ? /VERIFY_LINE_LIMIT = 4096/.test(source) : /maxBuffer/.test(source), `${label} bounds retained output`);
   const declared = Number(source.match(/^const DEFAULT_VERIFY_TIMEOUT_MS = (\d+);?$/m)[1]);
   assert.ok(declared >= 300000, `${label} default budget (${declared}ms) clears a several-minute suite`);
 }
@@ -95,15 +95,12 @@ assert.strictEqual(
   extractConst(cliSource, 'VERIFY_TIMEOUT_ENV').match(/'([^']*)'/)[1],
   'the plugin and the manual CLI read the same env override',
 );
-assert.strictEqual(
-  extractConst(pluginSource, 'VERIFY_MAX_BUFFER').match(/= (\d+ \* \d+ \* \d+)/)[1],
-  extractConst(cliSource, 'VERIFY_MAX_BUFFER').match(/= (\d+ \* \d+ \* \d+)/)[1],
-  'the plugin and the manual CLI allow the same output buffer',
-);
+// The plugin now streams output; the standalone CLI retains its 16 MiB
+// synchronous capture. Actual async runner contracts are exercised in 338.
 
 // Behavior: budget resolution.
 {
-  const context = loadVerificationLogic(pluginSource);
+  const context = loadVerificationLogic(cliSource);
   assert.strictEqual(context.verifyTimeoutMs(), 600000, 'unset env uses the default budget');
   for (const value of ['abc', '0', '-5', '']) {
     context.process.env.HARNESS_VERIFY_TIMEOUT_MS = value;
@@ -114,167 +111,9 @@ assert.strictEqual(
   assert.ok(!Number.isNaN(Number(context.verifyTimeoutMs())), 'the resolved budget is numeric');
 }
 
-// Behavior: outcome classification per failure kind.
-{
-  const context = loadVerificationLogic(pluginSource);
-  const calls = [];
-  context.execSync = (command, options) => {
-    calls.push({ command, options });
-    return '';
-  };
-  let result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.allPassed, true, 'a passing script verifies');
-  assert.strictEqual(result.results[0].outcome, 'pass', 'a passing script is classified pass');
-  assert.strictEqual(calls.length, 1, 'one command ran');
-  assert.strictEqual(calls[0].command, 'npm test', 'the test script is the command');
-  assert.strictEqual(calls[0].options.timeout, 600000, 'execSync receives the default budget');
-  assert.strictEqual(calls[0].options.maxBuffer, 16 * 1024 * 1024, 'execSync receives the raised output buffer');
-
-  context.process.env.HARNESS_VERIFY_TIMEOUT_MS = '120';
-  calls.length = 0;
-  context.execSync = (command, options) => {
-    calls.push({ command, options });
-    return '';
-  };
-  context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(calls[0].options.timeout, 120, 'execSync receives the overridden budget');
-
-  delete context.process.env.HARNESS_VERIFY_TIMEOUT_MS;
-  context.execSync = () => {
-    const error = new Error('spawnSync npm test ETIMEDOUT');
-    error.code = 'ETIMEDOUT';
-    error.signal = 'SIGTERM'; // the live execSync timeout kill shape
-    error.stdout = 'Harness OS - Self-Regression Test Suite\n';
-    throw error;
-  };
-  result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.allPassed, false, 'a timeout is a verification failure');
-  assert.strictEqual(result.results[0].timedOut, true, 'the timeout is recorded as a timeout');
-  assert.strictEqual(result.results[0].outcome, 'timeout', 'a timeout is classified as timeout, not test_failure');
-  assert.ok(/timed out after 600s/.test(result.results[0].error), 'the failure names the budget');
-  assert.ok(/HARNESS_VERIFY_TIMEOUT_MS/.test(result.results[0].error), 'the failure names the override');
-  assert.ok(/Self-Regression/.test(result.results[0].error), 'the failure keeps the last output for diagnosis');
-
-  context.process.env.HARNESS_VERIFY_TIMEOUT_MS = '50';
-  context.execSync = () => {
-    const started = Date.now();
-    while (Date.now() - started < 120) { /* simulate a kill at the budget without an error code */ }
-    const error = new Error('Command failed: npm test');
-    error.stdout = 'partial output';
-    throw error;
-  };
-  result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.results[0].timedOut, true, 'a kill at the budget is recognized without an ETIMEDOUT code');
-  assert.strictEqual(result.results[0].outcome, 'timeout', 'a kill at the budget is classified as timeout');
-
-  // P2: a signal is external termination; only budget expiry is a timeout.
-  context.process.env.HARNESS_VERIFY_TIMEOUT_MS = '600000';
-  context.execSync = () => {
-    const error = new Error('Command failed: npm test');
-    error.signal = 'SIGTERM'; // killed early by a user, host, or process manager
-    throw error; // status stays null: no verdict from the suite
-  };
-  result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.results[0].outcome, 'execution_error', 'an early external SIGTERM is an execution error, not a timeout');
-  assert.strictEqual(result.results[0].timedOut, false, 'an early external SIGTERM does not claim a timeout');
-  assert.strictEqual(result.results[0].signal, 'SIGTERM', 'the signal is retained for diagnostics');
-  assert.ok(/signal SIGTERM/.test(result.results[0].error), 'the execution error names the signal');
-
-  context.process.env.HARNESS_VERIFY_TIMEOUT_MS = '50';
-  context.execSync = () => {
-    const started = Date.now();
-    while (Date.now() - started < 80) { /* killed once the budget already elapsed, like execSync's own timeout */ }
-    const error = new Error('spawnSync npm test SIGTERM');
-    error.signal = 'SIGTERM';
-    throw error;
-  };
-  result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.results[0].outcome, 'timeout', 'a signal kill with no exit status once the budget elapsed is a timeout');
-
-  context.execSync = () => {
-    const started = Date.now();
-    while (Date.now() - started < 80) { /* slow, but it still produced a verdict */ }
-    const error = new Error('Command failed: npm test');
-    error.status = 1;
-    error.stdout = 'confirmed failure after a slow run';
-    throw error;
-  };
-  result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.results[0].outcome, 'test_failure', 'a real exit status stays a confirmed failure even past the budget');
-  assert.strictEqual(result.results[0].timedOut, false, 'a confirmed verdict is not a timeout');
-
-  context.process.env.HARNESS_VERIFY_TIMEOUT_MS = '600000';
-  context.execSync = () => {
-    const error = new Error('Command failed: npm test');
-    error.status = 1;
-    error.stdout = 'real assertion failure';
-    throw error;
-  };
-  result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.results[0].timedOut, false, 'a fast failure is not mislabeled as a timeout');
-  assert.strictEqual(result.results[0].outcome, 'test_failure', 'a non-zero exit is a confirmed test failure');
-  assert.strictEqual(result.results[0].error, 'real assertion failure', 'a fast failure reports its output');
-
-  context.execSync = () => {
-    const error = new Error('spawnSync npm test ENOBUFS');
-    error.code = 'ENOBUFS';
-    error.stdout = 'x'.repeat(300);
-    throw error;
-  };
-  result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.results[0].outcome, 'execution_error', 'ENOBUFS is an execution error, not a test failure');
-  assert.ok(/execution error: ENOBUFS/.test(result.results[0].error), 'the execution error names its cause');
-  assert.ok(!/timed out/.test(result.results[0].error), 'ENOBUFS is not reported as a timeout');
-
-  context.execSync = () => {
-    const error = new Error('spawnSync npm test ENOENT');
-    error.code = 'ENOENT';
-    throw error;
-  };
-  result = context.runVerification('/nonexistent-workspace');
-  assert.strictEqual(result.results[0].outcome, 'execution_error', 'a missing executable is an execution error');
-  assert.strictEqual(result.results[0].exitCode, null, 'a spawn-level error has no exit status');
-}
-
-// Behavior: test_failure evidence spans both streams in full (P1).
-{
-  const context = loadVerificationLogic(pluginSource);
-  context.process.env.HARNESS_VERIFY_TIMEOUT_MS = '600000';
-  context.execSync = () => {
-    const error = new Error('Command failed: npm test');
-    error.status = 1;
-    error.stdout = '> fixture@0.0.0 test\n> node fixture.js\n';
-    error.stderr = 'AssertionError [ERR_ASSERTION]: alpha boom\n';
-    throw error;
-  };
-  const bothStreams = context.runVerification('/nonexistent-workspace').results[0];
-  assert.strictEqual(bothStreams.outcome, 'test_failure', 'a both-stream failure is a confirmed test failure');
-  assert.ok(/alpha boom/.test(bothStreams.error), 'the user-visible diagnostic includes the stderr assertion');
-  assert.ok(/node fixture\.js/.test(bothStreams.error), 'the user-visible diagnostic keeps the stdout banner');
-  assert.ok(/alpha boom/.test(bothStreams.evidence), 'the kept evidence includes stderr');
-  assert.ok(/node fixture\.js/.test(bothStreams.evidence), 'the kept evidence includes stdout');
-
-  context.execSync = () => {
-    const error = new Error('Command failed: npm test');
-    error.status = 1;
-    error.stdout = `${'A'.repeat(2000)}TAIL-MARKER`;
-    throw error;
-  };
-  const longOutput = context.runVerification('/nonexistent-workspace').results[0];
-  assert.ok(longOutput.error.length < 2200, 'the user-visible diagnostic stays bounded');
-  assert.ok(/TAIL-MARKER/.test(longOutput.error), 'the trailing failure evidence survives diagnostic truncation');
-  assert.ok(/chars omitted/.test(longOutput.error), 'the elision is explicit in the diagnostic');
-  assert.ok(longOutput.evidence.length > 2000 && /TAIL-MARKER/.test(longOutput.evidence), 'the full evidence is retained untruncated for signature analysis');
-
-  context.execSync = () => {
-    const error = new Error('Command failed: npm test');
-    error.code = 'ENOBUFS';
-    error.stdout = `${'A'.repeat(2000)}TAIL-MARKER`;
-    throw error;
-  };
-  const spawnError = context.runVerification('/nonexistent-workspace').results[0];
-  assert.strictEqual(spawnError.evidence, undefined, 'only confirmed test failures keep full evidence');
-}
+// Plugin outcome, evidence and cleanup behavior is covered with real child
+// processes in mechanism-338-opencode-async.test.js, and downstream Rule-of-3
+// behavior remains in mechanism-39-opencode-verify-outcomes.test.js.
 
 // Behavior: the manual CLI mirror classifies the same way.
 {

package/docs/platform-capabilities.md

@@ -183,6 +183,14 @@ evidence supports project-scope `.js` loading and edit/verification state on
 OpenCode 1.18.31 (macOS): [live-host evidence](../benchmarks/results/live-host/opencode-2026-09-16/README.md). The
 retained snapshot predates #190 and remains historical evidence only. Current OpenCode removes the permanent post-reflection hard lock; only the third-failure reflection boundary pauses edits. Reflection was operator-seeded, then agent-rewritten, so do not promote this into a behavioral-effectiveness claim. `.mjs` auto-discovery is broken on this host version (issue #127). npm-package installation and behavioral V2-hook fire remain unverified.
 
+The plugin verification runner uses asynchronous child processes with a shared
+sequence budget and bounded streamed diagnostics. Deterministic mechanism
+coverage checks event-loop responsiveness, large-output verdicts, timeout/process
+cleanup, and concurrent edit/session reset handling. The standalone manual CLI
+retains its synchronous per-command runner. This change supplies no additional
+live-host evidence; the historical installation and behavior boundaries above
+remain unchanged.
+
 ## Official sources
 
 The matrix records the official documentation reviewed for each platform. Important source families include [OpenAI plugin packaging](https://developers.openai.com/plugins/build/plugins), [OpenAI public submission](https://developers.openai.com/plugins/deploy/submission), [Codex skills](https://learn.chatgpt.com/docs/build-skills), [Codex hooks](https://learn.chatgpt.com/docs/hooks), [Claude skills](https://code.claude.com/docs/en/skills), [OpenCode skills](https://opencode.ai/docs/skills), [GitHub Copilot Agent Skills](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/add-skills), [Cursor skills](https://cursor.com/docs/skills), [Continue rules](https://docs.continue.dev/customize/rules), and [Hermes skills](https://hermes-agent.nousresearch.com/docs/user-guide/features/skills).

package/harness-everything/SKILL.md

@@ -4,7 +4,7 @@ description: "Route software/project work through Harness: classify tier/topolog
 license: Apache-2.0
 metadata:
   author: Miya Daniel
-  version: 0.47.1
+  version: 0.47.3
 ---
 
 # Harness Everything

package/harness-everything/scripts/doc-check.js

@@ -247,11 +247,12 @@ function globToRegExp(pattern) {
   return new RegExp(`^${body}$`);
 }
 
-// Same file when both paths stat to one device and inode, so symlinks and hard links are caught.
+// Keep 64-bit device/file IDs exact (Windows IDs can exceed Number's safe
+// integer range), while still rejecting symlink and hard-link aliases.
 function sameFile(a, b) {
-  const x = fs.statSync(a);
-  const y = fs.statSync(b);
-  return x.ino > 0 && x.ino === y.ino && x.dev === y.dev;
+  const x = fs.statSync(a, { bigint: true });
+  const y = fs.statSync(b, { bigint: true });
+  return x.ino > 0n && x.ino === y.ino && x.dev === y.dev;
 }
 
 // Open without blocking (O_NONBLOCK keeps a FIFO from waiting for a writer), then require a regular file before reading.

package/hooks/scripts/lib/doc-reconciliation-provider.js

@@ -101,7 +101,7 @@ function isUncPathPrefix(token, command, index) {
   return token === '' && command[index] === '\\' && command[index + 1] === '\\';
 }
 // Canonical SHA-256 of harness-everything/scripts/doc-check.js (CRLF normalised to LF). Re-pin deliberately when that script changes.
-const SHIPPED_DOC_CHECK_DIGEST = '3ca386666d4dae9a98ffa27adf2f326bf171a783af90515375f0b328ca35f194';
+const SHIPPED_DOC_CHECK_DIGEST = '0e93f252b2350b145c34dcb5354aa3b4246f3c59e948a62c3a912305b355a475';
 // This install's own doc-check runtime, relative to hooks/scripts/lib: the source checkout or the packaged plugin.
 const DOC_CHECK_INSTALL_LAYOUTS = [
   ['..', '..', '..', 'harness-everything', 'scripts', 'doc-check.js'],

package/opencode-plugin/README.md

@@ -52,39 +52,37 @@ and on failure calls `client.session.prompt()` to push a synthetic follow-up
 message into the session - forcing the agent to keep working instead of
 actually stopping.
 
-The verification run has a budget, not a fixed 60s cap: a healthy self-regression
-suite takes around two minutes, so the default is 10 minutes and
-`HARNESS_VERIFY_TIMEOUT_MS` overrides it. Every run is classified as `pass`
-(exit 0), `test_failure` (a real non-zero exit), `timeout` (the budget actually
-expired: `ETIMEDOUT`, or a kill with no exit status once the budget elapsed) or
-`execution_error` (a spawn-level failure such as `ENOBUFS`/`ENOENT`, or an
-external signal such as `SIGTERM`/`SIGINT` before the budget - a signal alone
-is not a budget expiry).
-Only `test_failure` confirms a failing check and advances the Rule-of-3
-counter; `timeout` and `execution_error` are reported as uncompleted
-verifications and never impersonate repeated failures. A timeout is reported as
-`timed out after <n>s (set HARNESS_VERIFY_TIMEOUT_MS to raise it); last
-output: ...`, a spawn-level failure as `could not run to completion (execution
-error: ENOBUFS); output: ...` (an external signal names the signal).
-
-Known operational limits of the synchronous runner: `execSync` blocks the
-plugin's event loop for the duration of each script (worst case one budget per
-script), and captured output is buffered up to 16 MiB. A run whose output
-exceeds that buffer is classified as an `execution_error`, never as a failed
-check, so buffer pressure cannot flip the circuit breaker. Moving verification
-to an async `spawn` with streaming, bounded diagnostics is tracked as a
-follow-up in issue #338.
+The async runner streams stdout/stderr and gives the **whole ordered sequence**
+(`test`, `lint`, `build`) one wall-clock budget: 10 minutes by default, overridden
+by `HARNESS_VERIFY_TIMEOUT_MS`. It reports progress every 15 seconds in the
+plugin log. No later script starts after the budget expires. Each run is
+classified as `pass` (exit 0), `test_failure` (a real nonzero exit), `timeout`
+(the sequence deadline expired), or `execution_error` (spawn failure, external
+signal, or cancellation). Only `test_failure` advances the Rule-of-3 counter.
+
+| Resource | Bound / behavior |
+|---|---|
+| Retained output | First and last 2,000 characters per stream; the final diagnostic shares a 2,000-character budget across streams, plus labels/elision markers |
+| Failure identity | Incremental whitespace-normalized digest of both complete streams; first recognized test marker and suite context tracked separately |
+| Marker parsing | First 4,096 characters of each line; an overlong marker/context selects the complete-stream digest; ordinary long banners preserve later complete markers |
+| Timeout/cancel | Terminate the shell/npm/script process group, then force-kill after 150 ms on POSIX; Windows uses `taskkill /T /F` |
+| Concurrent lifecycle | Duplicate idle events share the active run; a code edit or session reset aborts it; an idle arriving during cancellation waits for cleanup before verifying the newer edits; the V2 event stream continues consuming cancellation events; disposal awaits active child cleanup; stale results cannot clear newer edits or recreate deleted state |
+
+The event loop remains available while scripts run. The output-retention limit
+never becomes a check verdict. Mechanism tests cover large output, shared
+budget, child cleanup and concurrent hooks; these are deterministic mechanism
+evidence, not a new live-host verification claim. Processes deliberately
+escaping the process group are outside this cleanup boundary.
 
 ### 3. Rule-of-3 zoom-out boundary (`tool.execute.before` + the verification observer above)
 
 Enforces Rule of 3 on repeated verification failures (same failing command +
 normalized failure evidence: the first test-level failing marker - TAP
 `not ok`, node:test/jest symbols, Jest test blocks, assertion lines - over the
-full captured stdout **and** stderr, with a preceding suite header such as
+streamed stdout **and** stderr, with a preceding suite header such as
 Jest's `FAIL <file>` carried as context so different tests in one file stay
 distinct; without a test-level marker the signature falls back to a digest of
-that full evidence, never to a suite header alone; the 2 KiB bound applies
-only to the user-visible diagnostic excerpt, never to signature analysis; not
+those complete streams, never to a suite header alone; the diagnostic retention bound does not limit the stream digest; not
 arbitrary tool failures, since
 `tool.execute.after` has no normalized success/failure field to key a
 signature on for tools in general):
@@ -146,17 +144,17 @@ installation surfaces and were not verified in this run.
 
 ## Manual verification CLI
 
-**File:** `hooks/verify.js` - the same "run available npm scripts" logic as
-`index.mjs`'s verification gate, kept as a standalone command:
+**File:** `hooks/verify.js` - a standalone "run available npm scripts" command:
 
 ```bash
 node opencode-plugin/hooks/verify.js
 ```
 
 Run it from the workspace under test, never from the Harness repo root - from
-there it would re-enter `npm test` from inside `npm test`. It is intentionally
-a separate copy of the logic, not a shared import, so `index.mjs` stays a
-single portable file for the installation step above.
+there it would re-enter `npm test` from inside `npm test`. The manual CLI retains its synchronous per-command budget and 16 MiB capture
+limit; the async sequence budget and streaming behavior above apply to the
+plugin only. Keeping its implementation separate preserves the single-file
+plugin installation contract.
 
 `plugin.json` records this standalone helper for repository parity checks; it
 does not configure opencode. The runtime module is `index.mjs`.