test(hooks): regression coverage for round 1 review fixes

9 new cases locking in the behavior added by the previous two
commits. Each was verified to fail before the fix and pass after.

Greptile — quote-aware depth counting:
  - blocks $(echo ")"; (npm run dev))
  - blocks (echo ")"; npm run dev)
  - allows $(echo "(npm run dev)") — () inside double-quoted body is literal

Greptile — brace groups:
  - blocks { npm run dev; }
  - blocks echo hi && { npm run dev; }
  - allows {npm run dev} — bash brace-group syntax requires a space after {

CodeRabbit — missing package-manager variants:
  - blocks yarn run dev (yarn 1.x convention)
  - blocks bun dev (bun bare form)

CodeRabbit nitpick — symmetric quote test:
  - blocks echo "$(npm run dev)" — double-quoted substitution still substitutes

The `{npm run dev}` allow case is intentional: bash treats `{` as
a reserved word only when followed by whitespace. The pre-fix code
already passed this through, but until now we never asserted it,
so a future change to brace handling could silently start blocking
literal `{npm` tokens.

docs/ECC-2.0-GA-ROADMAP.md

@@ -153,24 +153,6 @@ As of 2026-05-13:
   `/ecc-tools analyze --job status` now reads the #65 latest-result cache for
   the current PR head and posts a compact completed/blocked/not-run table with
   the next hosted job command, without queueing work or billing usage.
-- ECC-Tools PR #67 merged as `f20e6bec2b0bf49e4cc36e08b7285c795973b73d`
-  and made the hosted depth-plan check-run status-aware:
-  queued PR analysis now reads the #65/#66 latest-result cache when publishing
-  `ECC Tools / Hosted Depth Plan`, includes the latest hosted run status in
-  the plan table, and recommends the next unrun ready job before reruns.
-- ECC-Tools PR #68 merged as `2cde524b5ef8f34ab7bb1af973248fe4be4359f8`
-  and added deterministic hosted promotion readiness:
-  opened/synchronized PRs now publish a non-blocking
-  `ECC Tools / Hosted Promotion Readiness` check-run that compares changed
-  files against the checked-in evaluator/RAG corpus, warns on missing
-  hosted-job promotion evidence, and can be disabled with
-  `PR_HOSTED_PROMOTION_READINESS_CHECK_MODE=off`.
-- ECC-Tools PR #69 merged as `d0112dac7cef807ae27def41f057682ef0772cce`
-  and extended hosted promotion readiness with deterministic output scoring:
-  the check now reads cached completed hosted job results for the current PR
-  head, scores their artifacts and findings against evaluator/RAG corpus
-  expectations, and treats matching hosted artifacts as promotion evidence
-  before reporting a gap.
 - Handoff `ecc-supply-chain-audit-20260513-0645.md` under
   `~/.cluster-swarm/handoffs/`
   records the May 13 supply-chain sweep: no active lockfile/manifest hit for
@@ -404,16 +386,6 @@ As of 2026-05-13:
   `/ecc-tools analyze --job status`, summarizing completed, blocked, and
   not-yet-run hosted jobs for the PR head and recommending the next hosted job
   command.
-- ECC-Tools PR #67 feeds those cached results back into the hosted depth-plan
-  check-run so queued analysis recommends the next unrun ready hosted job from
-  cache state instead of repeating the static readiness order.
-- ECC-Tools PR #68 adds the first evaluator-backed hosted promotion gate:
-  opened/synchronized PRs get a non-blocking Hosted Promotion Readiness
-  check-run that turns the evaluator/RAG corpus into warnings when changed
-  files match fixture scenarios without their expected evidence artifacts.
-- ECC-Tools PR #69 extends that gate to score cached completed hosted job
-  outputs for the current PR head, so hosted artifacts can satisfy corpus
-  evidence expectations before the check reports a promotion gap.
 - ECC PR #1803 landed the contributor Quarkus handling branch after maintainer
   cleanup, current-`main` alignment, full local validation, and preservation of
   the author's removal of incomplete ja-JP and zh-CN Quarkus translations.
@@ -467,10 +439,10 @@ is not complete unless the evidence column exists and has been freshly verified.
 | Claude and Codex plugin publication | Contact/submission path with required artifacts and status | Publication readiness, naming matrix, and May 12 dry-run evidence document plugin validation, clean-checkout Claude tag/install smoke, and Codex marketplace CLI shape | Needs explicit approval for real tag/push and marketplace submission |
 | Articles, tweets, and announcements | X thread, LinkedIn copy, GitHub release copy, push checklist | Draft launch collateral exists under rc.1 release docs | Needs URL-backed refresh |
 | AgentShield enterprise iteration | Policy gates, SARIF, packs, provenance, corpus, HTML reports, exception lifecycle audit, baseline drift Action/CLI surfaces, evidence-pack redaction, harness adapter registry, enterprise research roadmap, supply-chain hardened release path, CI-safe baseline fingerprints, corpus accuracy recommendations, remediation workflow phases, env proxy hijack corpus coverage | PRs #53, #55-#64, #67-#69, and #78-#82 landed with test evidence; native PDF export deferred in favor of self-contained HTML plus print-to-PDF until explicit enterprise demand appears; `docs/architecture/agentshield-enterprise-research-roadmap.md` now has baseline drift, evidence-pack bundle, redaction, adapter-registry, supply-chain hardening, hashed baseline fingerprints, corpus accuracy recommendation, remediation workflow, and env proxy hijack corpus slices landed | Next hosted evidence-pack workflow depth |
-| ECC Tools next-level app | Billing audit, PR checks, deep analyzer, sync backlog, evaluator/RAG corpus, analysis-depth readiness, hosted execution planning, hosted CI diagnostics, hosted security evidence review, hosted harness compatibility audit, hosted reference-set evaluation, hosted AI routing/cost review, hosted team backlog routing, hosted depth-plan check-run, PR-comment hosted job dispatch, hosted job result history/check-runs, hosted result status command, status-aware depth-plan recommendations, hosted promotion readiness, hosted promotion output scoring | PRs #26-#43 plus #53-#69 landed with test evidence, including AgentShield evidence-pack gap routing, canonical bundle recognition, supply-chain signature gates, PR draft follow-up Linear tracking, evidence-backed/deep-ready repository classification, the `/api/analysis/depth-plan` hosted job plan, `/api/analysis/jobs/ci-diagnostics`, `/api/analysis/jobs/security-evidence-review`, `/api/analysis/jobs/harness-compatibility-audit`, `/api/analysis/jobs/reference-set-evaluation`, `/api/analysis/jobs/ai-routing-cost-review`, `/api/analysis/jobs/team-backlog-routing`, the `ECC Tools / Hosted Depth Plan` check-run, `/ecc-tools analyze --job ...` PR-comment dispatch, non-blocking per-hosted-job result check-runs backed by 30-day result cache records, `/ecc-tools analyze --job status` cache lookup, cache-aware next-job recommendations in the depth-plan check-run, the `ECC Tools / Hosted Promotion Readiness` corpus-backed PR check-run, and deterministic hosted-output scoring against cached completed job artifacts/findings | Next work is retrieval/model-backed hosted promotion after deterministic output scoring |
+| ECC Tools next-level app | Billing audit, PR checks, deep analyzer, sync backlog, evaluator/RAG corpus, analysis-depth readiness, hosted execution planning, hosted CI diagnostics, hosted security evidence review, hosted harness compatibility audit, hosted reference-set evaluation, hosted AI routing/cost review, hosted team backlog routing, hosted depth-plan check-run, PR-comment hosted job dispatch, hosted job result history/check-runs, hosted result status command | PRs #26-#43 plus #53-#66 landed with test evidence, including AgentShield evidence-pack gap routing, canonical bundle recognition, supply-chain signature gates, PR draft follow-up Linear tracking, evidence-backed/deep-ready repository classification, the `/api/analysis/depth-plan` hosted job plan, `/api/analysis/jobs/ci-diagnostics`, `/api/analysis/jobs/security-evidence-review`, `/api/analysis/jobs/harness-compatibility-audit`, `/api/analysis/jobs/reference-set-evaluation`, `/api/analysis/jobs/ai-routing-cost-review`, `/api/analysis/jobs/team-backlog-routing`, the `ECC Tools / Hosted Depth Plan` check-run, `/ecc-tools analyze --job ...` PR-comment dispatch, non-blocking per-hosted-job result check-runs backed by 30-day result cache records, and `/ecc-tools analyze --job status` cache lookup | Next work is evaluator-backed hosted promotion and status-aware depth-plan recommendations |
 | GitGuardian/Dependabot/CodeRabbit-style checks | Non-blocking taxonomy, deterministic follow-up checks, and local supply-chain gates | ECC-Tools risk taxonomy check plus follow-up signals landed, including Skill Quality, Deep Analyzer Evidence, Analyzer Corpus Evidence, RAG/Evaluator Evidence, PR Review/Salvage Evidence, and AgentShield evidence-pack evidence; #1846 added npm registry signature gates; #1848 added the supply-chain incident-response playbook and `pull_request_target` cache-poisoning validator guard; #1851 added the privileged checkout credential-persistence guard; AgentShield #78, JARVIS #13, and ECC-Tools #53 applied the same hardening outside trunk | Current supply-chain gate complete; deeper hosted review features remain future |
-| Harness-agnostic learning system | Audit, adapter matrix, observability, traces, promotion loop | Audit/adapters/observability gates plus `docs/architecture/evaluator-rag-prototype.md`, `examples/evaluator-rag-prototype/`, and ECC-Tools PR #40 define read-only stale-salvage, billing-readiness, CI-failure-diagnosis, harness-config-quality, AgentShield policy-exception, skill-quality evidence, deep-analyzer evidence, and RAG/evaluator comparison scenarios with trace, report, playbook, verifier, and predictive-check artifacts; ECC-Tools PRs #68/#69 now turn that corpus into a deterministic PR check-run gate with cached hosted-output scoring | Deterministic hosted PR check and cached output scoring integrated; hosted retrieval remains future |
-| Linear roadmap is detailed | Linear project status plus repo mirror | Repo mirror exists; issue creation was retried on 2026-05-12 and remains blocked by the workspace free issue limit; this May 13 sync adds ECC #1860, AgentShield #78-#82, JARVIS #13, ECC-Tools #53-#69, resolved queue/discussion counts, and Linear project status updates through ECC-Tools #69 | Needs recurring status updates after each merge batch |
+| Harness-agnostic learning system | Audit, adapter matrix, observability, traces, promotion loop | Audit/adapters/observability gates plus `docs/architecture/evaluator-rag-prototype.md`, `examples/evaluator-rag-prototype/`, and ECC-Tools PR #40 define read-only stale-salvage, billing-readiness, CI-failure-diagnosis, harness-config-quality, AgentShield policy-exception, skill-quality evidence, deep-analyzer evidence, and RAG/evaluator comparison scenarios with trace, report, playbook, verifier, and predictive-check artifacts | Local corpus complete; hosted integration remains future |
+| Linear roadmap is detailed | Linear project status plus repo mirror | Repo mirror exists; issue creation was retried on 2026-05-12 and remains blocked by the workspace free issue limit; this May 13 sync adds ECC #1860, AgentShield #78-#82, JARVIS #13, ECC-Tools #53-#66, resolved queue/discussion counts, and Linear project status updates through ECC-Tools #66 | Needs recurring status updates after each merge batch |
 | Flow separation and progress tracking | Flow lanes with owner artifacts and update cadence | This roadmap defines lanes below and `docs/architecture/progress-sync-contract.md` makes GitHub/Linear/handoff/roadmap sync part of the readiness gate | Active |
 | Realtime Linear sync | Project updates while issue limit is blocked; issues later | ECC-Tools #39 implements opt-in Linear API sync for deferred follow-up backlog items, and ECC-Tools #54 adds copy-ready PR drafts to that backlog when draft PR shells are not opened; `docs/architecture/progress-sync-contract.md` defines the local file-backed realtime boundary while issue capacity is blocked | Needs workspace capacity/config rollout |
 | Observability for self-use | Local readiness gate, traces, status snapshots, HUD/status contract, risk ledger, progress-sync contract | `npm run observability:ready` reports 21/21 | Complete for local gate |
@@ -489,9 +461,9 @@ repo evidence and merge commits.
 | Queue hygiene and salvage | GitHub PR/issue state, salvage ledger | Append ledger entries for any future stale closures | Every cleanup batch |
 | Release and publication | rc.1 release docs, publication readiness doc | Naming matrix and plugin submission/contact checklist | Before any tag |
 | Harness OS core | Audit, adapter matrix, observability docs, `ecc2/` | HUD/session-control acceptance spec | Weekly until GA |
-| Evaluation and RAG | Reference-set validation, harness audit, traces, ECC-Tools corpus | Read-only evaluator/RAG prototype plus stale-salvage, billing-readiness, CI-failure-diagnosis, harness-config-quality, AgentShield policy-exception, skill-quality evidence, deep-analyzer evidence, and RAG/evaluator comparison fixtures; ECC-Tools #68 publishes the corpus as a hosted promotion readiness check-run, and #69 scores cached hosted job outputs against the same corpus | Hosted retrieval/model-backed promotion plan |
+| Evaluation and RAG | Reference-set validation, harness audit, traces, ECC-Tools corpus | Read-only evaluator/RAG prototype plus stale-salvage, billing-readiness, CI-failure-diagnosis, harness-config-quality, AgentShield policy-exception, skill-quality evidence, deep-analyzer evidence, and RAG/evaluator comparison fixtures | Hosted retrieval/check-run automation plan |
 | AgentShield enterprise | AgentShield PR evidence and roadmap notes | Remediation workflow depth or corpus expansion follow-up | Next implementation batch |
-| ECC Tools app | ECC-Tools PR evidence, billing audit, risk taxonomy, evaluator/RAG corpus | ECC-Tools #53 published the supply-chain workflow hardening branch, #54 tracks copy-ready PR drafts in the Linear/project backlog, #55 classifies analysis-depth readiness, #56 exposes the hosted execution plan, #57 executes the first hosted CI diagnostics job, #58 executes the hosted security evidence review job, #59 executes the hosted harness compatibility audit, #60 executes the hosted reference-set evaluation, #61 executes the hosted AI routing/cost review, #62 executes hosted team backlog routing, #63 publishes the hosted depth-plan check-run, #64 dispatches hosted jobs from PR comments, #65 persists hosted result history/check-runs, #66 exposes hosted job status from PR comments, #67 makes depth-plan recommendations cache-aware, #68 publishes hosted promotion readiness from the evaluator/RAG corpus, and #69 scores cached hosted job outputs against that corpus; next work is retrieval/model-backed hosted promotion | Next implementation batch |
+| ECC Tools app | ECC-Tools PR evidence, billing audit, risk taxonomy, evaluator/RAG corpus | ECC-Tools #53 published the supply-chain workflow hardening branch, #54 tracks copy-ready PR drafts in the Linear/project backlog, #55 classifies analysis-depth readiness, #56 exposes the hosted execution plan, #57 executes the first hosted CI diagnostics job, #58 executes the hosted security evidence review job, #59 executes the hosted harness compatibility audit, #60 executes the hosted reference-set evaluation, #61 executes the hosted AI routing/cost review, #62 executes hosted team backlog routing, #63 publishes the hosted depth-plan check-run, and #64 dispatches hosted jobs from PR comments; next work is hosted result history/check-run summaries | Next implementation batch |
 | Linear progress | Linear project status updates, `docs/architecture/progress-sync-contract.md`, and this mirror | Status update with queue/evidence/missing gates | Every significant merge batch |
 
 The project status update should always include:
@@ -708,9 +680,9 @@ Acceptance:
    PR #82 expanded corpus coverage for env proxy hijacks and out-of-band
    exfiltration; and ECC-Tools PRs #42/#43 now route and recognize evidence
    packs. The next slice is hosted evidence-pack workflow depth.
-2. Plan retrieval/model-backed hosted promotion on top of the #69 deterministic
-   hosted output scoring contract, keeping vector/model judgment behind fixture
-   evaluation until the retrieval contract is stable.
+2. Feed the #66 status surface back into hosted depth-plan recommendations so
+   queued analysis can suggest the next unrun or newly blocked hosted job from
+   cached outcomes, not only static readiness.
 3. Enable/configure the merged Linear backlog sync path after workspace issue
    capacity clears or the Linear workspace is upgraded, then verify PR-draft
    salvage items land in the expected project.

scripts/hooks/pre-bash-dev-server-block.js

@@ -4,6 +4,10 @@
 const MAX_STDIN = 1024 * 1024;
 const path = require('path');
 const { splitShellSegments } = require('../lib/shell-split');
+const {
+  extractCommandSubstitutions,
+  extractSubshellGroups
+} = require('../lib/shell-substitution');
 
 const DEV_COMMAND_WORDS = new Set([
   'npm',
@@ -123,6 +127,8 @@ function getLeadingCommandWord(segment) {
       continue;
     }
 
+    if (token === '{' || token === '}') continue;
+
     if (/^[A-Za-z_][A-Za-z0-9_]*=.*/.test(token)) continue;
 
     const normalizedToken = normalizeCommandWord(token);
@@ -154,23 +160,55 @@ process.stdin.on('data', chunk => {
   }
 });
 
+const TMUX_LAUNCHER = /^\s*tmux\s+(new|new-session|new-window|split-window)\b/;
+const DEV_PATTERN = /\b(npm\s+run\s+dev|pnpm(?:\s+run)?\s+dev|yarn(?:\s+run)?\s+dev|bun(?:\s+run)?\s+dev)\b/;
+
+/**
+ * Collect every command-line segment we should evaluate. Returns the top-level
+ * segments first, then segments harvested from `$(...)` / backtick command
+ * substitutions and plain `(...)` subshell groups, recursively.
+ *
+ * Without this expansion the leading-command and dev-pattern check below only
+ * sees the outermost command, so wrappers like `$(npm run dev)` and
+ * `(npm run dev)` (which still spawn a dev server) sneak past.
+ */
+function collectCheckSegments(cmd) {
+  const segments = [...splitShellSegments(cmd)];
+  const queue = [cmd];
+  const seen = new Set();
+
+  while (queue.length) {
+    const current = queue.shift();
+    if (seen.has(current)) continue;
+    seen.add(current);
+
+    for (const body of extractCommandSubstitutions(current)) {
+      for (const seg of splitShellSegments(body)) segments.push(seg);
+      queue.push(body);
+    }
+    for (const body of extractSubshellGroups(current)) {
+      for (const seg of splitShellSegments(body)) segments.push(seg);
+      queue.push(body);
+    }
+  }
+
+  return segments;
+}
+
+function isBlockedDevSegment(segment) {
+  const commandWord = getLeadingCommandWord(segment);
+  if (!commandWord || !DEV_COMMAND_WORDS.has(commandWord)) return false;
+  return DEV_PATTERN.test(segment) && !TMUX_LAUNCHER.test(segment);
+}
+
 process.stdin.on('end', () => {
   try {
     const input = JSON.parse(raw);
     const cmd = String(input.tool_input?.command || '');
 
     if (process.platform !== 'win32') {
-      const segments = splitShellSegments(cmd);
-      const tmuxLauncher = /^\s*tmux\s+(new|new-session|new-window|split-window)\b/;
-      const devPattern = /\b(npm\s+run\s+dev|pnpm(?:\s+run)?\s+dev|yarn\s+dev|bun\s+run\s+dev)\b/;
-
-      const hasBlockedDev = segments.some(segment => {
-        const commandWord = getLeadingCommandWord(segment);
-        if (!commandWord || !DEV_COMMAND_WORDS.has(commandWord)) {
-          return false;
-        }
-        return devPattern.test(segment) && !tmuxLauncher.test(segment);
-      });
+      const segments = collectCheckSegments(cmd);
+      const hasBlockedDev = segments.some(isBlockedDevSegment);
 
       if (hasBlockedDev) {
         console.error('[Hook] BLOCKED: Dev server must run in tmux for log access');

scripts/lib/shell-substitution.js

@@ -0,0 +1,246 @@
+'use strict';
+
+/**
+ * Extract executable command-substitution bodies from a shell line.
+ *
+ * Single quotes are literal, so substitutions inside them are ignored;
+ * double quotes still permit substitutions, so those bodies are scanned
+ * before quoted text is stripped. Returns each substitution body plus
+ * any nested substitutions discovered recursively.
+ *
+ * Originally introduced in scripts/hooks/gateguard-fact-force.js
+ * (PR #1853 round 2). Extracted to a shared lib so other PreToolUse
+ * hooks that need the same "scan inside `$(...)` and backticks"
+ * behavior can reuse it without duplicating the parser.
+ *
+ * @param {string} input
+ * @returns {string[]}
+ */
+function extractCommandSubstitutions(input) {
+  const source = String(input || '');
+  const substitutions = [];
+  let inSingle = false;
+  let inDouble = false;
+
+  for (let i = 0; i < source.length; i++) {
+    const ch = source[i];
+    const prev = source[i - 1];
+
+    if (ch === '\\' && !inSingle) {
+      i += 1;
+      continue;
+    }
+
+    if (ch === "'" && !inDouble && prev !== '\\') {
+      inSingle = !inSingle;
+      continue;
+    }
+
+    if (ch === '"' && !inSingle && prev !== '\\') {
+      inDouble = !inDouble;
+      continue;
+    }
+
+    if (inSingle) {
+      continue;
+    }
+
+    if (ch === '`') {
+      let body = '';
+      i += 1;
+      while (i < source.length) {
+        const inner = source[i];
+        if (inner === '\\') {
+          body += inner;
+          if (i + 1 < source.length) {
+            body += source[i + 1];
+            i += 2;
+            continue;
+          }
+        }
+        if (inner === '`') {
+          break;
+        }
+        body += inner;
+        i += 1;
+      }
+      if (body.trim()) {
+        substitutions.push(body);
+        substitutions.push(...extractCommandSubstitutions(body));
+      }
+      continue;
+    }
+
+    if (ch === '$' && source[i + 1] === '(') {
+      let depth = 1;
+      let body = '';
+      let bodyInSingle = false;
+      let bodyInDouble = false;
+      i += 2;
+      while (i < source.length && depth > 0) {
+        const inner = source[i];
+        const innerPrev = source[i - 1];
+        if (inner === '\\' && !bodyInSingle) {
+          body += inner;
+          if (i + 1 < source.length) {
+            body += source[i + 1];
+            i += 2;
+            continue;
+          }
+        }
+        if (inner === "'" && !bodyInDouble && innerPrev !== '\\') {
+          bodyInSingle = !bodyInSingle;
+        } else if (inner === '"' && !bodyInSingle && innerPrev !== '\\') {
+          bodyInDouble = !bodyInDouble;
+        } else if (!bodyInSingle && !bodyInDouble) {
+          if (inner === '(') {
+            depth += 1;
+          } else if (inner === ')') {
+            depth -= 1;
+            if (depth === 0) {
+              break;
+            }
+          }
+        }
+        body += inner;
+        i += 1;
+      }
+      if (body.trim()) {
+        substitutions.push(body);
+        substitutions.push(...extractCommandSubstitutions(body));
+      }
+    }
+  }
+
+  return substitutions;
+}
+
+/**
+ * Extract bodies of plain `(...)` subshell groups.
+ *
+ * Bash treats `(npm run dev)` as a subshell that executes its contents, but
+ * the regex-light segment splitters used by our PreToolUse hooks don't peer
+ * inside those parens. This helper finds top-level `(...)` groups (skipping
+ * `$(...)` command substitutions and backticks, which `extractCommandSubstitutions`
+ * already covers) and returns each body, recursing for nested groups.
+ *
+ * Quote semantics:
+ * - Single quotes are literal: `'( ... )'` is a string, not a subshell.
+ * - Double quotes are literal *for parens*: `"( ... )"` is a string too —
+ *   bash only honors `$( )` inside double quotes, not bare `( )`.
+ *
+ * @param {string} input
+ * @returns {string[]}
+ */
+function extractSubshellGroups(input) {
+  const source = String(input || '');
+  const groups = [];
+  let inSingle = false;
+  let inDouble = false;
+
+  for (let i = 0; i < source.length; i++) {
+    const ch = source[i];
+    const prev = source[i - 1];
+
+    if (ch === '\\' && !inSingle) {
+      i += 1;
+      continue;
+    }
+
+    if (ch === "'" && !inDouble && prev !== '\\') {
+      inSingle = !inSingle;
+      continue;
+    }
+
+    if (ch === '"' && !inSingle && prev !== '\\') {
+      inDouble = !inDouble;
+      continue;
+    }
+
+    if (inSingle || inDouble) {
+      continue;
+    }
+
+    if (ch === '$' && source[i + 1] === '(') {
+      let depth = 1;
+      let skipInSingle = false;
+      let skipInDouble = false;
+      i += 2;
+      while (i < source.length && depth > 0) {
+        const inner = source[i];
+        const innerPrev = source[i - 1];
+        if (inner === '\\' && !skipInSingle) {
+          i += 2;
+          continue;
+        }
+        if (inner === "'" && !skipInDouble && innerPrev !== '\\') {
+          skipInSingle = !skipInSingle;
+        } else if (inner === '"' && !skipInSingle && innerPrev !== '\\') {
+          skipInDouble = !skipInDouble;
+        } else if (!skipInSingle && !skipInDouble) {
+          if (inner === '(') depth += 1;
+          else if (inner === ')') depth -= 1;
+        }
+        i += 1;
+      }
+      i -= 1;
+      continue;
+    }
+
+    if (ch === '`') {
+      i += 1;
+      while (i < source.length && source[i] !== '`') {
+        if (source[i] === '\\' && i + 1 < source.length) {
+          i += 2;
+          continue;
+        }
+        i += 1;
+      }
+      continue;
+    }
+
+    if (ch === '(') {
+      let depth = 1;
+      let body = '';
+      let bodyInSingle = false;
+      let bodyInDouble = false;
+      i += 1;
+      while (i < source.length && depth > 0) {
+        const inner = source[i];
+        const innerPrev = source[i - 1];
+        if (inner === '\\' && !bodyInSingle) {
+          body += inner;
+          if (i + 1 < source.length) {
+            body += source[i + 1];
+            i += 2;
+            continue;
+          }
+        }
+        if (inner === "'" && !bodyInDouble && innerPrev !== '\\') {
+          bodyInSingle = !bodyInSingle;
+        } else if (inner === '"' && !bodyInSingle && innerPrev !== '\\') {
+          bodyInDouble = !bodyInDouble;
+        } else if (!bodyInSingle && !bodyInDouble) {
+          if (inner === '(') {
+            depth += 1;
+          } else if (inner === ')') {
+            depth -= 1;
+            if (depth === 0) {
+              break;
+            }
+          }
+        }
+        body += inner;
+        i += 1;
+      }
+      if (body.trim()) {
+        groups.push(body);
+        groups.push(...extractSubshellGroups(body));
+      }
+    }
+  }
+
+  return groups;
+}
+
+module.exports = { extractCommandSubstitutions, extractSubshellGroups };

tests/docs/evaluator-rag-prototype.test.js

@@ -130,12 +130,12 @@ test('candidate playbook preserves stale-salvage operating rules', () => {
   }
 });
 
-test('roadmap points to the evaluator RAG prototype and hosted PR check', () => {
+test('roadmap points to the evaluator RAG prototype and keeps hosted integration open', () => {
   const roadmap = read('docs/ECC-2.0-GA-ROADMAP.md');
 
   assert.ok(roadmap.includes('docs/architecture/evaluator-rag-prototype.md'));
   assert.ok(roadmap.includes('examples/evaluator-rag-prototype/'));
-  assert.ok(roadmap.includes('Deterministic hosted PR check and cached output scoring integrated; hosted retrieval remains future'));
+  assert.ok(roadmap.includes('Local corpus complete; hosted integration remains future'));
 });
 
 test('billing readiness scenario rejects launch copy overclaims', () => {

tests/hooks/pre-bash-dev-server-block.test.js

@@ -89,6 +89,110 @@ function runTests() {
     assert.strictEqual(result.code, 0, `Expected exit code 0, got ${result.code}`);
   }) ? passed++ : failed++);
 
+  // --- Subshell bypass regression (issue: dev server slipped past via $(), ``, ()) ---
+
+  if (!isWindows) {
+    (test('blocks $(npm run dev) — command substitution', () => {
+      const result = runScript('$(npm run dev)');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+      assert.ok(result.stderr.includes('BLOCKED'), 'expected BLOCKED in stderr');
+    }) ? passed++ : failed++);
+
+    (test('blocks `npm run dev` — backtick substitution', () => {
+      const result = runScript('`npm run dev`');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks echo $(npm run dev) — substitution nested in argument', () => {
+      const result = runScript('echo $(npm run dev)');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks (npm run dev) — plain subshell group', () => {
+      const result = runScript('(npm run dev)');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks $(echo a; npm run dev) — substitution with sequenced segments', () => {
+      const result = runScript('$(echo a; npm run dev)');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks (pnpm dev) — plain subshell group with pnpm', () => {
+      const result = runScript('(pnpm dev)');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('allows tmux launcher inside subshell wrapping (exit code 0)', () => {
+      const result = runScript('(tmux new-session -d -s dev "npm run dev")');
+      assert.strictEqual(result.code, 0, `Expected exit code 0, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('allows single-quoted "(npm run dev)" — literal string, not a subshell', () => {
+      const result = runScript("git commit -m '(npm run dev)'");
+      assert.strictEqual(result.code, 0, `Expected exit code 0, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('allows double-quoted "(npm run dev)" — literal in double quotes (bash does not subshell)', () => {
+      const result = runScript('echo "(npm run dev)"');
+      assert.strictEqual(result.code, 0, `Expected exit code 0, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test("allows single-quoted '$(npm run dev)' — literal string, no substitution", () => {
+      const result = runScript("git commit -m '$(npm run dev) fix'");
+      assert.strictEqual(result.code, 0, `Expected exit code 0, got ${result.code}`);
+    }) ? passed++ : failed++);
+  }
+
+  // --- Round 1 review fixes (Greptile + CodeRabbit on PR #1889) ---
+
+  if (!isWindows) {
+    (test('blocks $(echo ")"; (npm run dev)) — quoted ) does not terminate $() early', () => {
+      const result = runScript('$(echo ")"; (npm run dev))');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks (echo ")"; npm run dev) — quoted ) does not terminate (...) early', () => {
+      const result = runScript('(echo ")"; npm run dev)');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('allows $(echo "(npm run dev)") — () inside double-quoted substitution body is literal', () => {
+      const result = runScript('$(echo "(npm run dev)")');
+      assert.strictEqual(result.code, 0, `Expected exit code 0, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks { npm run dev; } — brace group runs in current shell', () => {
+      const result = runScript('{ npm run dev; }');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks echo hi && { npm run dev; } — brace group after &&', () => {
+      const result = runScript('echo hi && { npm run dev; }');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('allows {npm run dev} — bash requires space after { to form a group', () => {
+      const result = runScript('{npm run dev}');
+      assert.strictEqual(result.code, 0, `Expected exit code 0, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks yarn run dev — yarn 1.x convention', () => {
+      const result = runScript('yarn run dev');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks bun dev — bun bare form', () => {
+      const result = runScript('bun dev');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+
+    (test('blocks "$(npm run dev)" — double-quoted substitution still substitutes', () => {
+      const result = runScript('echo "$(npm run dev)"');
+      assert.strictEqual(result.code, 2, `Expected exit code 2, got ${result.code}`);
+    }) ? passed++ : failed++);
+  }
+
   // --- Edge cases ---
 
   (test('empty/invalid input passes through (exit code 0)', () => {