@blamejs/exceptd-skills 0.13.25 → 0.13.27

package/CHANGELOG.md

@@ -1,12 +1,31 @@
 # Changelog
 
+## 0.13.27 — 2026-05-20
+
+`ci` next-step guidance now names the specific detected playbook(s) and surfaces pending jurisdiction obligations at the summary level — matching the regulatory-clock visibility a single `run` already gives.
+
+### Features
+
+- **`ci` FAIL Next steps names the actual detected playbook id.** Multi-playbook ci runs now print `Next steps (review the N detected finding(s) in <playbook-id>, ...)` with the real ids, and the `exceptd run <id> --format markdown` / `--format csaf-2.0` commands underneath use that id directly. Previously the output substituted a `<playbook>` placeholder that operators had to manually resolve against the per-playbook table above.
+- **`ci` surfaces pending jurisdiction obligations across all detected playbooks.** When at least one playbook lands `classification=detected`, the ci summary now prints `Pending jurisdiction obligations across detected playbook(s) (N) — clock starts on operator action:` grouped by `clock_start_event` — the same shape `run` emits, but aggregated across every playbook in the ci session. Operators gating a release no longer have to re-run each detected playbook individually to see the regulatory landscape.
+
+## 0.13.26 — 2026-05-20
+
+Internal hygiene. No operator-visible behavior change.
+
+### Internal
+
+- Stripped phase-residue version tags (`// v0.13.22 B5:`, `Pre-v0.13.23 the renderer ...`, `-v0_13_2X.test.js`) from comments and test filenames added in the v0.13.22–0.13.25 series. Authoritative version surfaces (`package.json` / `manifest.json` / `CHANGELOG.md ## X.Y.Z` headings / git tags / the CLI `version` verb) carry the version identifier; nowhere else.
+- New predeploy gate `Version-tag drift (no new phase residue)` (`scripts/check-version-tags.js`) refuses new `// v0.X.Y` / `Pre-v0.X.Y` comments and `*-v0_X_Y.test.js` filenames outside the authoritative surfaces. Baseline snapshot at `tests/.version-tag-baseline.json` captures pre-existing drift; refresh after an organic cleanup with `npm run check-version-tags:update`.
+- Gate count is now 17 (was 16).
+
 ## 0.13.25 — 2026-05-20
 
 Detected runs now surface pending jurisdiction obligations alongside the started ones — operators see the regulatory clock landscape at the same moment they see the finding, not after they remember to inspect the JSON.
 
 ### Features
 
-- **`run` human renderer surfaces pending jurisdiction obligations on detected runs.** Pre-v0.13.25 only `clock_started_at != null` obligations rendered; pending ones (waiting on `detect_confirmed` / `analyze_complete` / etc.) were invisible at the terminal even though the engine carried them in `phases.close.notification_actions`. Now a detected verdict prints `Pending jurisdiction obligations (N) — clock starts on operator action:` grouped by `clock_start_event`, then a `→ next: exceptd run <pb> --format csaf-2.0` pointer for the draft advisory + notification bodies. The grouping collapses one row per regulation into one row per clock-start event so the operator sees what action to take, not a wall of regulations.
+- **`run` human renderer surfaces pending jurisdiction obligations on detected runs.** Detected verdicts now print `Pending jurisdiction obligations (N) — clock starts on operator action:` grouped by `clock_start_event`, then a `→ next: exceptd run <pb> --format csaf-2.0` pointer for the draft advisory + notification bodies. Previously, only obligations whose `clock_started_at` was non-null surfaced at the terminal — pending ones (waiting on `detect_confirmed` / `analyze_complete` / etc.) were invisible even though the engine carried them in `phases.close.notification_actions`. Grouping by clock-start event collapses one row per regulation into one row per action the operator must take.
 
 ## 0.13.24 — 2026-05-20
 
@@ -19,14 +38,14 @@ Detected runs now surface pending jurisdiction obligations alongside the started
 
 ## 0.13.23 — 2026-05-19
 
-Continuation of the v0.13.22 UX pass: stage-by-stage next-step guidance so an operator (or an AI walking the workflow cold) never has to ask "what do I do now?"
+Stage-by-stage next-step guidance so an operator (or an AI walking the workflow cold) never has to ask "what do I do now?"
 
 ### Features
 
-- **`ci` human renderer emits a "Next steps" block per verdict.** BLOCKED → one `exceptd lint <playbook> -` command per blocked playbook plus the `--evidence <file>` re-run. NO_EVIDENCE → lint the first playbook + `ci --evidence-dir <dir>`. FAIL/detected → `run <playbook> --format markdown` / `--format csaf-2.0`. CLOCK_STARTED → `--format csaf-2.0` for the advisory draft. Pre-0.13.23 a blocked or no-evidence run printed only the reason — operators saw *why* they were stuck without the concrete command to unblock.
-- **`run` verdict line surfaces evidence_completeness.** Every successful run now shows `evidence: complete (13/13 indicators evaluated)` (or `partial` / `missing`) under the classification line. Distinguishes "ran every indicator and found nothing" from "couldn't evaluate, no evidence supplied" — pre-0.13.23 both states printed identically. When evidence is partial or missing, a `→ next: exceptd lint <playbook> -` pointer is appended.
-- **`run` attestation persistence is now visible.** Successful runs print `Attestation written: <full path>` followed by `exceptd attest verify <session-id>` and `exceptd attest diff <session-id>` so the operator knows where the JSON lives and how to verify or diff it. The persisted file path is also hoisted to the result envelope as `attestation_path`. Pre-0.13.23 the attestation went to `~/.exceptd/attestations/<repo>@<branch>/<session-id>/attestation.json` with zero indication in any output.
-- **`run` remediation prose matches the verdict.** Pre-0.13.23 every classification printed `Recommended remediation: <id>` — misleading on `not_detected` and `inconclusive` runs, where there is nothing to remediate. Now non-detect runs print `Remediation path (informational — verdict=<x>, no action required now): <id>`; detected runs are unchanged.
+- **`ci` human renderer emits a "Next steps" block per verdict.** BLOCKED → one `exceptd lint <playbook> -` command per blocked playbook plus the `--evidence <file>` re-run. NO_EVIDENCE → lint the first playbook + `ci --evidence-dir <dir>`. FAIL/detected → `run <playbook> --format markdown` / `--format csaf-2.0`. CLOCK_STARTED → `--format csaf-2.0` for the advisory draft. Previously, a blocked or no-evidence run printed only the reason — operators saw *why* they were stuck without the concrete command to unblock.
+- **`run` verdict line surfaces evidence_completeness.** Every successful run now shows `evidence: complete (13/13 indicators evaluated)` (or `partial` / `missing`) under the classification line. Distinguishes "ran every indicator and found nothing" from "couldn't evaluate, no evidence supplied" — previously, both states printed identically. When evidence is partial or missing, a `→ next: exceptd lint <playbook> -` pointer is appended.
+- **`run` attestation persistence is now visible.** Successful runs print `Attestation written: <full path>` followed by `exceptd attest verify <session-id>` and `exceptd attest diff <session-id>` so the operator knows where the JSON lives and how to verify or diff it. The persisted file path is also hoisted to the result envelope as `attestation_path`. Previously, the attestation went to `~/.exceptd/attestations/<repo>@<branch>/<session-id>/attestation.json` with zero indication in any output.
+- **`run` remediation prose matches the verdict.** Non-detect runs now print `Remediation path (informational — verdict=<x>, no action required now): <id>` instead of the unconditional `Recommended remediation: <id>` that previously fired on every classification — misleading on `not_detected` and `inconclusive` runs, where there is nothing to remediate. Detected runs unchanged.
 
 ### Bugs
 
@@ -39,7 +58,7 @@ Continuation of the v0.13.22 UX pass: stage-by-stage next-step guidance so an op
 
 ### Features
 
-- **Human-readable `ci` output by default.** Pre-0.13.22 the default `ci` output was 1000+ lines of indented JSON on every run. Now the default is a one-screen digest: verdict line, per-playbook table (id / verdict / rwep / evidence-completeness / top-finding), session-level warnings, scope-selection rules, framework gap rollup, and fail reasons. Pass `--json` or `--pretty` to get the structured body for automation.
+- **Human-readable `ci` output by default.** The default `ci` output is now a one-screen digest: verdict line, per-playbook table (id / verdict / rwep / evidence-completeness / top-finding), session-level warnings, scope-selection rules, framework gap rollup, and fail reasons. Previously, the default was 1000+ lines of indented JSON on every run. Pass `--json` or `--pretty` to get the structured body for automation.
 - **Per-result hoisted summary fields.** Every `run()` result now carries `verdict`, `rwep_score`, `top_finding`, `summary_line`, and `evidence_completeness` (one of `complete` / `partial` / `missing` / `unknown` / `not-evaluated`) at the top level. Machine-readable consumers no longer walk `phases.analyze.rwep.adjusted` and `phases.detect.classification` separately to extract the headline numbers.
 - **`indicators_evaluated` + `indicators_known` per result.** Surface how many of the playbook's known indicators were actually exercised by the operator's evidence, so a result that returns `verdict=inconclusive` with `indicators_evaluated=0` is distinguishable from one that evaluated every indicator and found no hits.
 - **Session-level warning de-duplication.** `ci` runs that span N playbooks no longer emit the same `bundle_publisher_unclaimed` warning N times. The summary now carries `runtime_warnings` and `runtime_warnings_count` with one entry per unique (kind, reason) across the session.

package/bin/exceptd.js

@@ -2289,14 +2289,12 @@ function cmdLint(runner, args, runOpts, pretty) {
       });
     }
   } else {
-    // v0.13.23 — nested submission with artifacts but no signal_overrides
-    // also lands every indicator on inconclusive. The flat-shape branch
-    // already surfaced this; the nested-shape silence was the workflow gap
-    // operators hit: they fill in lint's per-artifact guidance, run, and
-    // get an opaque "every indicator inconclusive" result. Surface the
-    // signal_overrides shape explicitly so the operator/AI knows the
-    // next step is to set `signal_overrides[<indicator-id>] = "hit" |
-    // "miss" | "inconclusive"` per indicator they investigated.
+    // Nested submission with artifacts but no signal_overrides lands
+    // every indicator on inconclusive — same trapdoor the flat-shape
+    // branch above surfaces. Detect() needs signal_overrides (or a
+    // verdict override) to drive an indicator hit/miss; artifact
+    // presence alone is not enough. Surface the JSON shape explicitly
+    // so the operator/AI knows what to populate.
     const verdictClass = submission.verdict?.classification;
     const verdictWillDrive = verdictClass === "clean" || verdictClass === "not_detected" || verdictClass === "detected" || verdictClass === "inconclusive";
     const normalizedHasOverrides = Object.keys(normalized.signal_overrides || {}).length > 0;
@@ -2949,9 +2947,9 @@ function cmdRun(runner, args, runOpts, pretty) {
       result.prior_session_id = persistResult.prior_session_id;
       result.overwrote_at = persistResult.overwrote_at;
     }
-    // v0.13.23 — surface the attestation file path so the human renderer
-    // can echo it and the next-step guidance (attest verify <sid>) lands
-    // on an artifact the operator can actually find.
+    // Surface the persisted file path on the result so the human
+    // renderer can echo it and the attest verify / attest diff hint
+    // lands on an artifact the operator can actually find.
     if (persistResult.attestation_path) {
       result.attestation_path = persistResult.attestation_path;
     }
@@ -3178,12 +3176,11 @@ function cmdRun(runner, args, runOpts, pretty) {
     const top = rwep?.threshold?.escalate ?? "n/a";
     const verdictIcon = cls === "detected" ? "[!! DETECTED]" : cls === "inconclusive" ? "[i  INCONCLUSIVE]" : "[ok]";
     lines.push(`\n${verdictIcon}  classification=${cls}  RWEP ${adj}/${top}${adj !== base ? ` (Δ${adj - base} from operator evidence)` : " (catalog baseline)"}  blast_radius=${obj.phases?.analyze?.blast_radius_score ?? "n/a"}/5`);
-    // v0.13.23 — surface evidence_completeness on the verdict line so
-    // operators distinguish "ran every indicator and found nothing"
-    // (evidence=complete) from "couldn't evaluate, no evidence supplied"
-    // (evidence=missing). Pre-v0.13.23 the two states printed identically
-    // — a not_detected run with zero evidence looked the same as one
-    // with a fully-populated submission.
+    // Surface evidence_completeness on the verdict line so operators
+    // distinguish "ran every indicator and found nothing"
+    // (evidence=complete) from "couldn't evaluate, no evidence
+    // supplied" (evidence=missing) — without this they look identical
+    // at the terminal.
     if (obj.evidence_completeness && obj.indicators_known != null) {
       const ev = obj.evidence_completeness;
       const ke = obj.indicators_evaluated ?? 0;
@@ -3231,13 +3228,13 @@ function cmdRun(runner, args, runOpts, pretty) {
       lines.push(`\nIndicators that fired (${hits.length}):`);
       for (const i of hits.slice(0, 8)) lines.push(`  ${i.id}  (${i.confidence}${i.deterministic ? "/deterministic" : ""})`);
     }
-    // v0.13.23 — selected_remediation is informational on non-detect
-    // runs (validate() always picks the highest-priority remediation
-    // path as a "what would you do IF you found something" anchor).
-    // Rendering it as "Recommended remediation:" on a not_detected /
-    // inconclusive verdict misleads operators into thinking action is
-    // required. Tag it conditionally so the operator-facing prose
-    // matches the verdict.
+    // selected_remediation is informational on non-detect runs:
+    // validate() always picks the highest-priority remediation path
+    // as a "what you'd do IF you found something" anchor, even when
+    // classification is not_detected / inconclusive. Tag the prose
+    // conditionally so the label matches the verdict — labeling it
+    // "Recommended remediation:" on a not_detected run misleads
+    // operators into thinking action is required.
     const rem = obj.phases?.validate?.selected_remediation;
     if (rem) {
       if (cls === "detected") {
@@ -3247,12 +3244,11 @@ function cmdRun(runner, args, runOpts, pretty) {
       }
       lines.push(`  ${rem.description?.slice(0, 200) || ""}`);
     }
-    // v0.13.25 — surface BOTH started and pending notification clocks.
-    // Pre-v0.13.25 the renderer hid pending obligations on detected
-    // runs, so an operator finishing a detected scan never saw that
-    // jurisdiction X / regulation Y was waiting on `detect_confirmed`
-    // or `analyze_complete` to start a 24h / 72h / 720h clock. That
-    // misses the entire reason regulators care about this tool.
+    // Surface BOTH started and pending notification clocks on detected
+    // runs. The detection IS the regulatory event for the obligations
+    // exceptd tracks — pending obligations waiting on detect_confirmed
+    // / analyze_complete are exactly what the operator needs to see
+    // before taking the action that starts the clock.
     const allNotif = obj.phases?.close?.notification_actions || [];
     const startedNotif = allNotif.filter(n => n.clock_started_at);
     const pendingNotif = allNotif.filter(n => !n.clock_started_at);
@@ -3278,12 +3274,11 @@ function cmdRun(runner, args, runOpts, pretty) {
     const feeds = obj.phases?.close?.feeds_into || [];
     if (feeds.length) lines.push(`\nNext playbooks suggested: ${feeds.join(", ")}`);
 
-    // v0.13.23 — tell the operator WHERE the attestation went and HOW
-    // to verify/diff it. Pre-v0.13.23 a successful run silently wrote
+    // Tell the operator WHERE the attestation went and HOW to verify
+    // / diff it. Without this, the attestation goes to
     // ~/.exceptd/attestations/<repo>@<branch>/<sid>/attestation.json
-    // with zero indication to the operator. The next-time-they-asked-
-    // about-this-run lookup ("attest verify <sid>") then failed with
-    // "no session dir" because they were in a different cwd.
+    // and a follow-up `attest verify <sid>` from a different cwd
+    // fails with "no session dir" because the lookup is cwd-tagged.
     if (obj.attestation_path) {
       lines.push(`\nAttestation written: ${obj.attestation_path}`);
       lines.push(`  exceptd attest verify ${obj.session_id}     # tamper check`);
@@ -3845,11 +3840,9 @@ function persistAttestation(args) {
 
     try {
       writeAttestation(null, null, "wx");
-      // v0.13.23 — surface the absolute path so the caller can echo it
-      // in the human renderer. Pre-v0.13.23 the attestation was silently
-      // written to ~/.exceptd/attestations/<repo>@<branch>/<session-id>/
-      // and the operator had no way to find it short of grepping the
-      // filesystem.
+      // Return the absolute path so the caller can echo it in the
+      // human renderer — operators need to know where the file went
+      // for follow-up `attest verify` / `attest diff` calls.
       return { ok: true, prior_session_id: null, overwrote_at: null, attestation_path: filePath };
     } catch (eExcl) {
       if (eExcl.code !== "EEXIST") throw eExcl;
@@ -4620,10 +4613,9 @@ function cmdReattest(runner, args, runOpts, pretty) {
     // on-disk artifact without re-deriving the filename.
     replay_persisted: replayPersisted,
   }, pretty, (obj) => {
-    // v0.13.24 — human renderer for `attest diff` (reattest path).
-    // Pre-v0.13.24 the only output was JSON, so an operator asking "did
-    // anything change since the last run?" had to parse the envelope to
-    // get the one-line answer they wanted.
+    // Human renderer for `attest diff` (reattest path) — one-screen
+    // answer to "did anything change since the last run?" so the
+    // operator doesn't have to parse the JSON envelope.
     const lines = [];
     lines.push(`attest diff: ${obj.session_id} (${obj.playbook_id})`);
     const icon = obj.status === "unchanged" ? "[ok]" : "[i  DRIFTED]";
@@ -4962,10 +4954,9 @@ function cmdAttest(runner, args, runOpts, pretty) {
       body.replay_tamper = true;
       body.warnings = ["one or more replay records failed Ed25519 verification — audit-trail corruption suspected, regenerate via reattest"];
     }
-    // v0.13.24 — human renderer for `attest verify`. Pre-v0.13.24 the
-    // only output was a JSON envelope, even at the terminal. The whole
-    // point of `attest verify` is "did anyone tamper with my evidence
-    // since I ran it?" — that question deserves a one-line answer.
+    // Human renderer for `attest verify` — one-line answer to "did
+    // anyone tamper with my evidence since I ran it?" so the operator
+    // doesn't have to parse the JSON envelope.
     emit(body, pretty, (obj) => {
       const lines = [];
       lines.push(`attest verify: ${obj.session_id}`);
@@ -6989,12 +6980,12 @@ function cmdCi(runner, args, runOpts, pretty) {
     clock_started_reasons: clockStartedReasons,
   };
 
-  // v0.13.22 B5: Each `run()` call independently surfaces session-level
-  // runtime conditions (e.g. bundle_publisher_unclaimed) into its own
-  // phases.analyze.runtime_errors. On a ci run that spans 9 playbooks, the
-  // operator saw the same warning 9 times. Dedupe across results by
-  // (kind, reason) so a session-level condition surfaces once at the ci
-  // summary level — operators read one row, not N copies.
+  // Each `run()` call independently surfaces session-level runtime
+  // conditions (e.g. bundle_publisher_unclaimed) into its own
+  // phases.analyze.runtime_errors. On a ci run spanning N playbooks
+  // the same warning would otherwise appear N times. Dedupe across
+  // results by (kind, reason) so session-level conditions surface
+  // once at the summary level, not once per playbook.
   const warningSeen = new Set();
   const runtimeWarningsDedup = [];
   for (const r of results) {
@@ -7013,11 +7004,11 @@ function cmdCi(runner, args, runOpts, pretty) {
   summary.runtime_warnings = runtimeWarningsDedup;
   summary.runtime_warnings_count = runtimeWarningsDedup.length;
 
-  // v0.13.22 B8 (transparency): document why each playbook was selected.
-  // --scope <s> always adds cross-cutting; --scope code on a repo with a
-  // lockfile also adds sbom. The selection rule was buried in code; surface
-  // it in the summary so operators reading the JSON / pretty trailer can
-  // see what was scoped vs. auto-included.
+  // Document why each playbook was selected. --scope <s> always adds
+  // cross-cutting; --scope code on a repo with a lockfile also adds
+  // sbom. The selection rule is otherwise buried in code; surface it
+  // in the summary so operators reading the output can see what was
+  // scoped vs. auto-included.
   if (scope) {
     summary.scope_request = scope;
     summary.scope_inclusion_rules = [
@@ -7071,16 +7062,14 @@ function cmdCi(runner, args, runOpts, pretty) {
     );
     return;
   } else {
-    // v0.13.22 B3+B4+B6: human renderer for `ci` default output. Pre-0.13.22
-    // the only output was indented JSON (or compact JSON when not TTY) —
-    // operators running `exceptd ci` at the terminal saw 1000+ lines of JSON
-    // for a 9-playbook scan and had to grep for the verdict by hand.
-    //
-    // Renderer shape (one screen for a typical 9-playbook scope):
-    //   - one verdict line (PASS/FAIL/BLOCKED + counts)
-    //   - per-playbook row: id | verdict | rwep | evidence | top_finding
-    //   - deduped session-level runtime warnings (B5)
-    //   - scope inclusion rules (B8 transparency) when --scope was used
+    // Human renderer for `ci` default output. Shape (one screen for a
+    // typical 9-playbook scope):
+    //   - verdict line (PASS / FAIL / BLOCKED / CLOCK_STARTED / NO_EVIDENCE + counts)
+    //   - per-playbook table (id | verdict | rwep | evidence | top_finding)
+    //   - session-level runtime warnings (deduped by kind+reason)
+    //   - scope inclusion rules when --scope was used
+    //   - jurisdiction-clock + framework-gap rollup
+    //   - fail reasons + per-verdict next-step block
     //   - footer pointing at --json / --format for the structured body
     emit({ verb: "ci", session_id: sessionId, playbooks_run: ids, summary, results }, pretty, (obj) => {
       const s = obj.summary;
@@ -7168,9 +7157,9 @@ function cmdCi(runner, args, runOpts, pretty) {
         for (const r of s.fail_reasons) lines.push(`  - ${r}`);
       }
 
-      // v0.13.23 — Next-step guidance. An operator (or an AI walking the
-      // workflow cold) reading the ci output should never have to ask
-      // "what do I do now?" The verdict dictates the next move:
+      // Next-step guidance, keyed on verdict. An operator reading ci
+      // output should never have to ask "what do I do now?" — the
+      // verdict dictates the next move:
       //   BLOCKED        → operator must supply evidence asserting the
       //                    halted preconditions; `exceptd lint <pb> -`
       //                    emits the exact JSON paths to fill in.
@@ -7194,17 +7183,54 @@ function cmdCi(runner, args, runOpts, pretty) {
         lines.push(lintCmd(firstId));
         lines.push(`  exceptd ci --scope <type> --evidence-dir <dir>  # gate again with real submissions`);
       } else if (s.verdict === "FAIL") {
-        // codex P2 (PR #63): FAIL can fire in two distinct shapes —
+        // FAIL fires in two distinct shapes:
         //   (a) at least one playbook classification=detected → s.detected > 0
         //   (b) inconclusive playbook(s) whose rwep_delta (operator
         //       evidence) crossed the cap → s.detected stays at 0
-        // Pre-fix this branch was keyed on `s.detected > 0`, so the
-        // shape-(b) FAIL exited with no "Next steps" footer at all,
-        // contradicting the v0.13.23 stage-by-stage promise.
+        // Both shapes need actionable Next-step guidance; key on the
+        // shape, not on `s.detected > 0` alone.
         if (s.detected > 0) {
-          lines.push(`\nNext steps (review the ${s.detected} detected finding(s)):`);
-          lines.push(`  exceptd run <playbook> --format markdown    # operator-readable digest`);
-          lines.push(`  exceptd run <playbook> --format csaf-2.0    # advisory bundle for downstream`);
+          // Name the specific detected playbook ids so the operator
+          // can copy-paste rather than substitute `<playbook>`. When
+          // multiple playbooks land detected, emit one row per id
+          // for each format so operators don't miss follow-up for the
+          // playbooks beyond detectedIds[0].
+          const detectedIds = (obj.results || [])
+            .filter(r => r && r.ok !== false && r.phases?.detect?.classification === "detected")
+            .map(r => r.playbook_id)
+            .filter(Boolean);
+          const ids = detectedIds.length ? detectedIds : ["<playbook>"];
+          lines.push(`\nNext steps (review the ${s.detected} detected finding(s) in ${detectedIds.join(", ") || "<playbook>"}):`);
+          for (const id of ids) {
+            lines.push(`  exceptd run ${id} --format markdown    # operator-readable digest`);
+          }
+          for (const id of ids) {
+            lines.push(`  exceptd run ${id} --format csaf-2.0    # advisory bundle for downstream`);
+          }
+
+          // Surface pending jurisdiction obligations across all
+          // detected playbooks at the ci summary level — operators
+          // running ci to gate a PR / a release deserve the same
+          // regulatory-clock visibility a single `run` would give them.
+          const pendingByEvent = {};
+          let pendingTotal = 0;
+          for (const r of obj.results || []) {
+            if (r?.phases?.detect?.classification !== "detected") continue;
+            const notif = r?.phases?.close?.notification_actions || [];
+            for (const n of notif) {
+              if (n.clock_started_at) continue;
+              const ev = n.clock_start_event || "unspecified";
+              if (!pendingByEvent[ev]) pendingByEvent[ev] = new Set();
+              pendingByEvent[ev].add(`${n.jurisdiction || "?"}/${n.regulation || "?"} (${n.window_hours || "?"}h)`);
+              pendingTotal++;
+            }
+          }
+          if (pendingTotal > 0) {
+            lines.push(`\nPending jurisdiction obligations across detected playbook(s) (${pendingTotal}) — clock starts on operator action:`);
+            for (const [ev, refs] of Object.entries(pendingByEvent)) {
+              lines.push(`  on ${ev}:  ${[...refs].join(", ")}`);
+            }
+          }
         } else {
           // Operator evidence pushed RWEP across --max-rwep cap on an
           // otherwise-inconclusive run. The fix is to review which

package/data/_indexes/_meta.json

@@ -1,10 +1,10 @@
 {
   "schema_version": "1.1.0",
-  "generated_at": "2026-05-20T00:48:54.230Z",
+  "generated_at": "2026-05-20T03:47:07.431Z",
   "generator": "scripts/build-indexes.js",
   "source_count": 54,
   "source_hashes": {
-    "manifest.json": "56140e0d1624f8c83333f27759e21815daf57a8face69d7aff3eb88d087f27b1",
+    "manifest.json": "9e9f77b63bb9cf8a34e235fe103e3e71efa558f2802fd563f73b18dc8f516d99",
     "data/atlas-ttps.json": "d296c1d3e71807c9279b731f047e57796e85137f186586743a8cdad214b408f9",
     "data/attack-techniques.json": "49b6010b317edd219def135171ea8f3b1bbf1e00e9c5a08bf7237215ff54e2c3",
     "data/cve-catalog.json": "a09c83af3f9679a7ea73935726a1ff9de2cab94b4ab6321fc017fc147747d7c3",

package/lib/playbook-runner.js

@@ -3058,13 +3058,13 @@ function run(playbookId, directiveId, agentSubmission = {}, runOpts = {}) {
   }
   const pre = preflight(playbook, { ...runOpts, precondition_checks: mergedPCs });
   if (!pre.ok) {
-    // v0.13.22 (B7 fix): blocked results MUST carry playbook_id +
-    // directive_id at the result root. Pre-v0.13.22 a `ci --scope code`
-    // run produced 9 results where the 3 blocked rows had no
-    // playbook_id, so operators iterating results[] couldn't tell
-    // which playbook produced which preflight failure without joining
-    // against playbooks_run[] by index. Also include `verdict:"blocked"`
-    // and a one-line `summary_line` for B2 + B6 flat-result-shape fixes.
+    // Blocked results MUST carry playbook_id + directive_id at the
+    // result root, the same as successful results. Without this,
+    // consumers iterating results[] can't identify which playbook
+    // produced a preflight failure without joining against
+    // playbooks_run[] by array index. `verdict:"blocked"` +
+    // `summary_line` keep the flat result-envelope shape consistent
+    // across both branches.
     const summaryLine = `${playbookId}: blocked at preflight (${pre.blocked_by || 'unknown'}) — ${pre.reason || ''}`.slice(0, 240);
     return {
       ok: false,
@@ -3226,18 +3226,16 @@ function run(playbookId, directiveId, agentSubmission = {}, runOpts = {}) {
       }))
       .digest('hex');
 
-    // v0.13.22 — hoist the operator-facing summary to the result root
-    // (B2). Pre-v0.13.22 a `ci` consumer had to spelunk into
-    // phases.detect.classification + phases.analyze.rwep.adjusted to
-    // find the answer; the most common ops question ("did this
-    // playbook detect anything?") needs a one-line answer at the top.
+    // Hoist the operator-facing summary to the result root so a `ci`
+    // consumer does not have to walk into phases.detect.classification
+    // + phases.analyze.rwep.adjusted separately to answer the most
+    // common ops question ("did this playbook detect anything?").
     //
     // Verdict source: phases.detect.classification is the canonical
-    // verdict signal — one of detected / not_detected / inconclusive /
-    // pending / skipped. validate() does NOT carry a `verdict` field;
-    // an earlier draft of this hoist read phases.validate.verdict and
-    // therefore degraded every non-blocked result to "inconclusive"
-    // (codex P1 on PR #62).
+    // signal — one of detected / not_detected / inconclusive / pending
+    // / skipped. validate() does NOT carry a `verdict` field, so
+    // sourcing the hoist from phases.validate.verdict would degrade
+    // every non-blocked result to the fallback "inconclusive".
     const verdict = (phases.detect && phases.detect.classification) || 'inconclusive';
     const rwepScore = phases.analyze && phases.analyze.rwep && typeof phases.analyze.rwep.adjusted === 'number'
       ? phases.analyze.rwep.adjusted : null;
@@ -3253,10 +3251,12 @@ function run(playbookId, directiveId, agentSubmission = {}, runOpts = {}) {
         && phases.detect.classification !== 'pending') {
       topFinding = phases.detect.classification;
     }
-    // Evidence completeness (B9): the indicators-evaluated count vs
-    // the indicators-known count distinguishes "ran fully and found
-    // nothing" from "couldn't actually evaluate". Pre-v0.13.22 the two
-    // states looked identical at the result-root level.
+    // Evidence completeness: indicators-evaluated vs indicators-known
+    // distinguishes "ran fully and found nothing" from "couldn't
+    // actually evaluate". Without this, the two states look identical
+    // at the result-root level — a not_detected verdict with zero
+    // indicators-evaluated reads the same as one with every indicator
+    // evaluated and miss.
     const indicatorsKnown = (playbook.phases && playbook.phases.detect && playbook.phases.detect.indicators)
       ? playbook.phases.detect.indicators.length : null;
     const indicatorsEvaluated = phases.detect && typeof phases.detect.indicators_evaluated_count === 'number'

package/manifest.json

@@ -1,6 +1,6 @@
 {
   "name": "exceptd-security",
-  "version": "0.13.25",
+  "version": "0.13.27",
   "description": "AI security skills grounded in mid-2026 threat reality, not stale framework documentation",
   "homepage": "https://exceptd.com",
   "license": "Apache-2.0",
@@ -53,7 +53,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "lXhZgoIrrVloO3XaTvo/43AxZn4mwErstd7DR0O/oVhD3AOGODM4HqrageYEou9WKOdMEGP5mJNTjJsXdP5NDA==",
-      "signed_at": "2026-05-20T00:48:05.112Z",
+      "signed_at": "2026-05-20T03:47:06.133Z",
       "cwe_refs": [
         "CWE-125",
         "CWE-362",
@@ -117,7 +117,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "vSVqu4wBm+d68ujZmM6Rto/HzViCkE0gPUcv/MYE/bjFiqamf/s0On4kTOo1KIveV9cOwYNxiItaGEWlVkRFDg==",
-      "signed_at": "2026-05-20T00:48:05.114Z",
+      "signed_at": "2026-05-20T03:47:06.135Z",
       "cwe_refs": [
         "CWE-1039",
         "CWE-1426",
@@ -180,7 +180,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "RIgXKvolQjgJdnlrDnVOd90IOY1B7VHHZD/YJQRzouL+wUeOLclPrdK/EgEuFyiu7lR4bi+Pl6aGB9G9tOxYCQ==",
-      "signed_at": "2026-05-20T00:48:05.114Z",
+      "signed_at": "2026-05-20T03:47:06.135Z",
       "cwe_refs": [
         "CWE-22",
         "CWE-345",
@@ -226,7 +226,7 @@
       "framework_gaps": [],
       "last_threat_review": "2026-05-01",
       "signature": "RYOxeq/o3uTwTWq4H7RcdH2Aclg9UyCERfUH9Frwkzncsowg7LgxpaEDc3swTCv73HMEGbU8wVbXguZ4JxHUCQ==",
-      "signed_at": "2026-05-20T00:48:05.114Z"
+      "signed_at": "2026-05-20T03:47:06.136Z"
     },
     {
       "name": "compliance-theater",
@@ -257,7 +257,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "DneJCPKCPcoe6nQ82XptqSqNfSRdt1orKaO+o7K36YCciDrzwJb+1BuBLusPDtpcdDaGY0y0e+AqiTYJklhBAQ==",
-      "signed_at": "2026-05-20T00:48:05.115Z"
+      "signed_at": "2026-05-20T03:47:06.136Z"
     },
     {
       "name": "exploit-scoring",
@@ -286,7 +286,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "NA1hoQycvQhSUoG5rwlXX0mOVmGxoXRVezkELGEA2nZOdGis4gXkHT3O6Sfw7zxE4JuMrsCb65TEeOWk9WEPDg==",
-      "signed_at": "2026-05-20T00:48:05.115Z"
+      "signed_at": "2026-05-20T03:47:06.137Z"
     },
     {
       "name": "rag-pipeline-security",
@@ -323,7 +323,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "XgrzcA2brPhXrSTxrcLnJec0OpgGYJBoSTUlJ10UdePHffxqb9LTVGnfbmEk1ykQifXREZexui2bG7X/+eFfCQ==",
-      "signed_at": "2026-05-20T00:48:05.116Z",
+      "signed_at": "2026-05-20T03:47:06.137Z",
       "cwe_refs": [
         "CWE-1395",
         "CWE-1426"
@@ -380,7 +380,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "9+hZlZOqZdeACUmamQk66L5levZhhwnFXuYRhdT6Mce99eQaKT7wNfWq12hXQztkRcVRKaFH+a01zwJQwsRQCA==",
-      "signed_at": "2026-05-20T00:48:05.116Z",
+      "signed_at": "2026-05-20T03:47:06.138Z",
       "d3fend_refs": [
         "D3-CA",
         "D3-CSPP",
@@ -415,7 +415,7 @@
       "framework_gaps": [],
       "last_threat_review": "2026-05-01",
       "signature": "ciqhVloMWWXEigPZvvwoV2c54tEqsDqsoc+sS/mNTFFJk2H+tz2+XUrgfEPRuYw0FeyNB6/+27pL2NpKHzUqAg==",
-      "signed_at": "2026-05-20T00:48:05.116Z",
+      "signed_at": "2026-05-20T03:47:06.138Z",
       "cwe_refs": [
         "CWE-1188"
       ]
@@ -443,7 +443,7 @@
       "framework_gaps": [],
       "last_threat_review": "2026-05-01",
       "signature": "xiHAhhdufm9hCKU8PLiPE0MX65ej2F4OZwtlWLGLCiie9/km+Kiqbt192LcMvr94v83C98pb9wIaqFsFWft6AQ==",
-      "signed_at": "2026-05-20T00:48:05.117Z"
+      "signed_at": "2026-05-20T03:47:06.138Z"
     },
     {
       "name": "global-grc",
@@ -475,7 +475,7 @@
       "framework_gaps": [],
       "last_threat_review": "2026-05-01",
       "signature": "oYsSk35N2Uzq7MRofACykylcVwkgPhI4luWZ14vmQT+gUKLyZiKVOUJbe1+7lGl6BYPRN0sUDQ0f7S5Eu5w2Ag==",
-      "signed_at": "2026-05-20T00:48:05.117Z"
+      "signed_at": "2026-05-20T03:47:06.139Z"
     },
     {
       "name": "zeroday-gap-learn",
@@ -502,7 +502,7 @@
       "framework_gaps": [],
       "last_threat_review": "2026-05-01",
       "signature": "igRqYyU1unRFH40BsPyAR62SPrk8QZv8dPGb8S9O9EvLCNOZAzm3t+HdT/NKqzWHwrpomOzkkkyLfYI/0qTUDA==",
-      "signed_at": "2026-05-20T00:48:05.117Z"
+      "signed_at": "2026-05-20T03:47:06.139Z"
     },
     {
       "name": "pqc-first",
@@ -554,7 +554,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "vhc3wuQEro/86s1ro2b/KakUXg8QVnySYTBqA7ebzv9oeR2HYO5bvGEJp3oOHWtL37JDqcCAHYadSN/qxIyCCA==",
-      "signed_at": "2026-05-20T00:48:05.118Z",
+      "signed_at": "2026-05-20T03:47:06.139Z",
       "cwe_refs": [
         "CWE-327"
       ],
@@ -601,7 +601,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "MS35nWm8djfJGn4OOoT0JKJ2aO+Dkbb6wOOWJYvNZlRKT3UGA59o2gxg1JOnD20hb/RwxtkmCujhl2tuYSR+AQ==",
-      "signed_at": "2026-05-20T00:48:05.118Z"
+      "signed_at": "2026-05-20T03:47:06.140Z"
     },
     {
       "name": "security-maturity-tiers",
@@ -638,7 +638,7 @@
       ],
       "last_threat_review": "2026-05-01",
       "signature": "8Px1s2lDj10/Q6erwEQlXgUHM1+OTruUR8qAHPX7Oo3k/l69N6P9sm0PsafS9wDFtj9l5C/OiLiFgzMlMt6vBw==",
-      "signed_at": "2026-05-20T00:48:05.118Z",
+      "signed_at": "2026-05-20T03:47:06.140Z",
       "cwe_refs": [
         "CWE-1188"
       ]
@@ -673,7 +673,7 @@
       "framework_gaps": [],
       "last_threat_review": "2026-05-11",
       "signature": "WAu5fRirzSOcnnZsTx2d/JJZwa/LPpXCi+31qATTGLmoNuhyy81k3ooPe9kCM3E0CLMtvTePg9DagYqBninZDQ==",
-      "signed_at": "2026-05-20T00:48:05.119Z"
+      "signed_at": "2026-05-20T03:47:06.140Z"
     },
     {
       "name": "attack-surface-pentest",
@@ -744,7 +744,7 @@
         "PTES revision incorporating AI-surface enumeration"
       ],
       "signature": "7eEwCXFd9pDKUw7yCUbRJSjfzozE44dwwwemCQUPm8JBPztLltibD9bL/RszSbYyCrYJmVb5Drncz2cGe62gCw==",
-      "signed_at": "2026-05-20T00:48:05.119Z"
+      "signed_at": "2026-05-20T03:47:06.141Z"
     },
     {
       "name": "fuzz-testing-strategy",
@@ -804,7 +804,7 @@
         "OSS-Fuzz-Gen / AI-assisted harness generation becoming the default expectation for OSS maintainers"
       ],
       "signature": "Z7ypCUnXx8JpLtgxxB6RHNi39w74AmrGY1N4ofAGCXhkuM2EaFVm1AU0dvl9UQ1bVLfHKEDGqMO/TwlIY7RABg==",
-      "signed_at": "2026-05-20T00:48:05.119Z"
+      "signed_at": "2026-05-20T03:47:06.141Z"
     },
     {
       "name": "dlp-gap-analysis",
@@ -879,7 +879,7 @@
         "Quebec Law 25, India DPDPA, KSA PDPL enforcement actions naming AI-tool prompt data as in-scope personal information"
       ],
       "signature": "fgxG344JGYBWWWwFXZ1IzGipWKP7EyBhrsvsbsb0CCGXfv/MvNHVNI6G0zQddCsWX1JeQbhZT3Vk8v1uJKDTDA==",
-      "signed_at": "2026-05-20T00:48:05.120Z"
+      "signed_at": "2026-05-20T03:47:06.141Z"
     },
     {
       "name": "supply-chain-integrity",
@@ -956,7 +956,7 @@
         "OpenSSF model-signing — emerging Sigstore-based signing standard for ML model weights; track for production adoption"
       ],
       "signature": "pcLrM98A3vUSZRjwNAk0aZ9umvOwB41XCLLsCOy/IebB2F/06oIrGUKkMHtHwm4pTVPShMMcKdZQQ3jz30FnCg==",
-      "signed_at": "2026-05-20T00:48:05.120Z"
+      "signed_at": "2026-05-20T03:47:06.142Z"
     },
     {
       "name": "defensive-countermeasure-mapping",
@@ -1013,7 +1013,7 @@
       ],
       "last_threat_review": "2026-05-11",
       "signature": "gqF8eU3VBrZhO2WnlcqKa7wm1d2mmWtvpbmx0kNCgHojNV+qEt+Ij84RO6bZvaUqhfYPWizWL79Fa4DL0curAQ==",
-      "signed_at": "2026-05-20T00:48:05.120Z"
+      "signed_at": "2026-05-20T03:47:06.142Z"
     },
     {
       "name": "identity-assurance",
@@ -1080,7 +1080,7 @@
       "d3fend_refs": [],
       "last_threat_review": "2026-05-11",
       "signature": "Wv5hGMeHjlaQK1zwicVCA7AvdKgJBgvcjdpGM9Ywahh9tagAKhbkOjybowDQZzu7OZ3bDkbh6pBYc1Sdwr6NAA==",
-      "signed_at": "2026-05-20T00:48:05.121Z"
+      "signed_at": "2026-05-20T03:47:06.142Z"
     },
     {
       "name": "ot-ics-security",
@@ -1136,7 +1136,7 @@
       "d3fend_refs": [],
       "last_threat_review": "2026-05-11",
       "signature": "8t5qKHd3yWi57dvG36YQkLN/X9bQWqtEiYjay4IfSmqhJpM/xXPaQVKNGz3wscrO8OLKUZ0OaX7Mj5kzpgBKBQ==",
-      "signed_at": "2026-05-20T00:48:05.121Z"
+      "signed_at": "2026-05-20T03:47:06.143Z"
     },
     {
       "name": "coordinated-vuln-disclosure",
@@ -1188,7 +1188,7 @@
         "NYDFS 23 NYCRR 500 amendments potentially adding explicit CVD program requirements"
       ],
       "signature": "GDGt4UPqBa04PjlpSmpyihGzd3OgfBN7jaAK5tfwp+LRSs3ygKOdbeivUCCHNagTY1hE6hG2Ou40ADfBFuXeAg==",
-      "signed_at": "2026-05-20T00:48:05.121Z"
+      "signed_at": "2026-05-20T03:47:06.143Z"
     },
     {
       "name": "threat-modeling-methodology",
@@ -1238,7 +1238,7 @@
         "PASTA v2 updates incorporating AI/ML application threats"
       ],
       "signature": "rFBpOQEJUPpl+v88Lw/WqVJRhTl80vy0VbPAbzQj3Q0suJRRrJg368I9uKu5LXIBKFDvKxnGIcIzbGg9NUtaCA==",
-      "signed_at": "2026-05-20T00:48:05.122Z"
+      "signed_at": "2026-05-20T03:47:06.143Z"
     },
     {
       "name": "webapp-security",
@@ -1312,7 +1312,7 @@
       "d3fend_refs": [],
       "last_threat_review": "2026-05-11",
       "signature": "ux85YI4t2mVHOyt744Yin1HHy+z11JIFygjKfFfQOBBl5QVV3A267jeIy7utix85irMcpZm/T3yx/ooqiK2tBA==",
-      "signed_at": "2026-05-20T00:48:05.122Z"
+      "signed_at": "2026-05-20T03:47:06.144Z"
     },
     {
       "name": "ai-risk-management",
@@ -1362,7 +1362,7 @@
       "d3fend_refs": [],
       "last_threat_review": "2026-05-11",
       "signature": "IIXnkZ5ZNqFwOto5KfytADTLLZLoyXNZACD1ORZ40P1HUAQxe6u2uyXFzzsfuob4Uy06jNkRGr2FFgCphUH1Cw==",
-      "signed_at": "2026-05-20T00:48:05.122Z"
+      "signed_at": "2026-05-20T03:47:06.144Z"
     },
     {
       "name": "sector-healthcare",
@@ -1422,7 +1422,7 @@
       "d3fend_refs": [],
       "last_threat_review": "2026-05-11",
       "signature": "AhF9KF8ZBlDteciV+F8IBSmFVYCvQOn44GmD4rZjgLoPxfIv/QE1/vSkK32zyqDKtHWkLSXExbkkPkxA/V6dDw==",
-      "signed_at": "2026-05-20T00:48:05.123Z"
+      "signed_at": "2026-05-20T03:47:06.145Z"
     },
     {
       "name": "sector-financial",
@@ -1503,7 +1503,7 @@
         "TIBER-EU framework v2.0 alignment with DORA TLPT RTS (JC 2024/40); cross-recognition with CBEST and iCAST"
       ],
       "signature": "HQgZvb4ReziEz5rNFr8i/O8/rJEZR+iHRROT7m/D2QUqhrcNISPkYXENsUZlG8xapzy/Ik92ehkseyj4hdmhCQ==",
-      "signed_at": "2026-05-20T00:48:05.123Z"
+      "signed_at": "2026-05-20T03:47:06.145Z"
     },
     {
       "name": "sector-federal-government",
@@ -1572,7 +1572,7 @@
         "Australia PSPF 2024 revision and ISM quarterly updates — track for Essential Eight Maturity Level requirements for federal entities"
       ],
       "signature": "linxmsXZiOYtcs71sSWgGCrvb8xQfmxmtTY5PRvZJ0/8FgJulo0tQtejzexYG775s7XhjAmGsDP238BQTQ8ADA==",
-      "signed_at": "2026-05-20T00:48:05.124Z"
+      "signed_at": "2026-05-20T03:47:06.145Z"
     },
     {
       "name": "sector-energy",
@@ -1637,7 +1637,7 @@
         "ICS-CERT advisory feed (https://www.cisa.gov/news-events/cybersecurity-advisories/ics-advisories) for vendor CVEs in Siemens, Rockwell, Schneider Electric, ABB, GE Vernova, Hitachi Energy, AVEVA / OSIsoft PI"
       ],
       "signature": "JjBfc0ovta560Clk0x3QGRM5osFJDwcvpy3rT7QEGdCIL827jzE8QCow1C8deXq+4JhY2sA/d7/8IsxikdlkCg==",
-      "signed_at": "2026-05-20T00:48:05.124Z"
+      "signed_at": "2026-05-20T03:47:06.146Z"
     },
     {
       "name": "sector-telecom",
@@ -1723,7 +1723,7 @@
         "O-RAN SFG / WG11 security specifications"
       ],
       "signature": "JWVxKFoKrbX4d+Tko1d4OBdwyg25MfFFKn4CT6E/CzH+YwnU3T6Y76uBQIKg3+gIGTvPduqyvQwQQ5FxKDuPBw==",
-      "signed_at": "2026-05-20T00:48:05.124Z"
+      "signed_at": "2026-05-20T03:47:06.146Z"
     },
     {
       "name": "api-security",
@@ -1792,7 +1792,7 @@
       "d3fend_refs": [],
       "last_threat_review": "2026-05-11",
       "signature": "BmCRCestWqr55+fCynEhtAl5NWLT+xLTkpwS0Icp3SaoZOw/ce3Y6TtqjHRSKn4CBJq7YDiLRWxmhO3MStvOAA==",
-      "signed_at": "2026-05-20T00:48:05.125Z"
+      "signed_at": "2026-05-20T03:47:06.146Z"
     },
     {
       "name": "cloud-security",
@@ -1873,7 +1873,7 @@
         "CISA KEV additions for cloud-control-plane CVEs (IMDSv1 abuses, federation token mishandling, cross-tenant boundary failures); CISA Cybersecurity Advisories for cross-cloud advisories"
       ],
       "signature": "/DV3pmZwrRySrk1OCbyI+0BQESacjupJfUX3eC2NGtXuYOBro0vndIP+z27heFxumnjU3a9sfla7/U9X+pqnDw==",
-      "signed_at": "2026-05-20T00:48:05.125Z"
+      "signed_at": "2026-05-20T03:47:06.147Z"
     },
     {
       "name": "container-runtime-security",
@@ -1935,7 +1935,7 @@
       "d3fend_refs": [],
       "last_threat_review": "2026-05-11",
       "signature": "E2UGSf9ATyYgzBr8uM/0ubOUmDqo1jVA7f9mVxv6LHfWGCNuQNXDyuNou9VAmUCeeXEeUYIi3AFjXkJqpOkxDA==",
-      "signed_at": "2026-05-20T00:48:05.125Z"
+      "signed_at": "2026-05-20T03:47:06.147Z"
     },
     {
       "name": "mlops-security",
@@ -2006,7 +2006,7 @@
         "MITRE ATLAS v5.6.0 (released February 2026) shipped the AML.T0010 sub-technique expansion this forecast tracked plus new techniques (\"Publish Poisoned AI Agent Tool\", \"Escape to Host\"); inventory now 16 tactics, 84 techniques, 56 sub-techniques. Forward watch: ATLAS v5.5 / v6.0 — track next-cadence updates to agentic-AI TTPs and MLOps-pipeline-specific techniques"
       ],
       "signature": "BGNE6ZQWBA1LmsUFe8tU0L67iGDSrFqiuqaZD2f1KqfcyqqzQfMs9PWNHFzxxaJmXeKlm87eU8lgELF0bX+RBA==",
-      "signed_at": "2026-05-20T00:48:05.126Z"
+      "signed_at": "2026-05-20T03:47:06.148Z"
     },
     {
       "name": "incident-response-playbook",
@@ -2068,7 +2068,7 @@
         "NYDFS 23 NYCRR 500.17 amendments tightening ransom-payment 24h disclosure operationalization"
       ],
       "signature": "FkZQerh3VHVJAwIcCktDyMRh5KE2+Em/i0ek8zEz7JG/PXtQx8ujHWTh3VjZbOLhPNtdB2qxgXOIAYIofaVOAQ==",
-      "signed_at": "2026-05-20T00:48:05.126Z"
+      "signed_at": "2026-05-20T03:47:06.148Z"
     },
     {
       "name": "ransomware-response",
@@ -2148,7 +2148,7 @@
       ],
       "last_threat_review": "2026-05-15",
       "signature": "n3UToNuN3A1HgLvcuqmIx8vrZY71+r/79waK92jG+rSX4uYOzkmxMUpROrE5K9bDwMezNBHdjWv8Uul6zugyDQ==",
-      "signed_at": "2026-05-20T00:48:05.127Z"
+      "signed_at": "2026-05-20T03:47:06.148Z"
     },
     {
       "name": "email-security-anti-phishing",
@@ -2201,7 +2201,7 @@
       "d3fend_refs": [],
       "last_threat_review": "2026-05-11",
       "signature": "rK+WnuS+9tqEABmwc0jO/PEmxcLjG1/tmUb897HsClQeKzf+TQOlwBE+OsbtuKxpjYNwur62Xxs3TxObkwm8Cw==",
-      "signed_at": "2026-05-20T00:48:05.127Z"
+      "signed_at": "2026-05-20T03:47:06.149Z"
     },
     {
       "name": "age-gates-child-safety",
@@ -2269,7 +2269,7 @@
         "US state adult-site age-verification laws — 19+ states by mid-2026 (TX HB 18 upheld by SCOTUS June 2025 in Free Speech Coalition v. Paxton); track ongoing challenges in remaining states"
       ],
       "signature": "+OO0RhQ303RJV7kaH38IuZpLeQbapep6Ds4Re/WEZu0FHBwKSlwvF7jbtP7KQ57xldJYn/xZm2jaszyOacMfDg==",
-      "signed_at": "2026-05-20T00:48:05.127Z"
+      "signed_at": "2026-05-20T03:47:06.149Z"
     },
     {
       "name": "cloud-iam-incident",
@@ -2349,7 +2349,7 @@
       ],
       "last_threat_review": "2026-05-15",
       "signature": "e/kij7GtKaytROyIj7V5RH+FC9WtmVFzrmG2kIlNDNn29ep/CRNlIQKwXLpzo/81AIf634pmdr1qy/+vwIuUDA==",
-      "signed_at": "2026-05-20T00:48:05.128Z"
+      "signed_at": "2026-05-20T03:47:06.149Z"
     },
     {
       "name": "idp-incident-response",
@@ -2430,11 +2430,11 @@
       ],
       "last_threat_review": "2026-05-15",
       "signature": "ew9Kglc9fAZzbn0ZIfGP7WSK/j4eV2VhSvpy+s5bEfNEVYIMa2kZjnGBapgUsyGDLes9H9K2ovjQyX17+GKiBw==",
-      "signed_at": "2026-05-20T00:48:05.128Z"
+      "signed_at": "2026-05-20T03:47:06.150Z"
     }
   ],
   "manifest_signature": {
     "algorithm": "Ed25519",
-    "signature_base64": "wn3gVy+dARzm5jSMzCYVWnNZaqzV3XpvKSmZ49+W4W+AwBZGoW6lo+MNPaQ8bKKpLgmwx94EMOnGhgyFM5dvAQ=="
+    "signature_base64": "clr4Flw551fOYatT6y2GZqejMax68GStrCb24cYycm5/YhRugCF/X7cjrzy+FrjlKOz0jEVPmRsg+vE+GjeTAg=="
   }
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@blamejs/exceptd-skills",
-  "version": "0.13.25",
+  "version": "0.13.27",
   "description": "AI security skills grounded in mid-2026 threat reality, not stale framework documentation. 42 skills, 10 catalogs (312 CVEs / 171 CWEs / 805 ATT&CK + ICS / 170 ATLAS / 468 D3FEND / 7476 RFCs), 34 jurisdictions, 10-class catalog gap detector + budget gate, real XML parser + canonical-form diff + content-pattern regression detection, Ed25519-signed.",
   "keywords": [
     "ai-security",
@@ -101,6 +101,8 @@
     "audit-catalog-gaps": "node scripts/audit-catalog-gaps.js --pretty",
     "audit-catalog-gaps:json": "node scripts/audit-catalog-gaps.js",
     "audit-catalog-gaps:strict": "node scripts/audit-catalog-gaps.js --pretty --strict",
+    "check-version-tags": "node scripts/check-version-tags.js",
+    "check-version-tags:update": "node scripts/check-version-tags.js --update-baseline",
     "predeploy": "node scripts/predeploy.js",
     "diff-coverage": "node scripts/check-test-coverage.js",
     "prepublishOnly": "node -e \"if(process.env.EXCEPTD_SKIP_PREPUBLISH_PREDEPLOY!=='1'){const r=require('child_process').spawnSync(process.execPath,['scripts/predeploy.js'],{stdio:'inherit'});if(r.status){process.exit(r.status)}}\" && node lib/validate-package.js",

package/sbom.cdx.json

@@ -1,22 +1,22 @@
 {
   "bomFormat": "CycloneDX",
   "specVersion": "1.6",
-  "serialNumber": "urn:uuid:9721d2eb-b479-4a91-8e53-f6c07d9d9333",
+  "serialNumber": "urn:uuid:e25943d6-47a9-4837-b3db-bff8760f7682",
   "version": 1,
   "metadata": {
-    "timestamp": "2106-05-08T22:44:59.000Z",
+    "timestamp": "2146-05-04T13:21:26.000Z",
     "tools": [
       {
         "vendor": "blamejs",
         "name": "scripts/refresh-sbom.js",
-        "version": "0.13.25"
+        "version": "0.13.27"
       }
     ],
     "component": {
-      "bom-ref": "pkg:npm/@blamejs/exceptd-skills@0.13.25",
+      "bom-ref": "pkg:npm/@blamejs/exceptd-skills@0.13.27",
       "type": "application",
       "name": "@blamejs/exceptd-skills",
-      "version": "0.13.25",
+      "version": "0.13.27",
       "description": "AI security skills grounded in mid-2026 threat reality, not stale framework documentation. 42 skills, 10 catalogs (312 CVEs / 171 CWEs / 805 ATT&CK + ICS / 170 ATLAS / 468 D3FEND / 7476 RFCs), 34 jurisdictions, 10-class catalog gap detector + budget gate, real XML parser + canonical-form diff + content-pattern regression detection, Ed25519-signed.",
       "licenses": [
         {
@@ -25,17 +25,17 @@
           }
         }
       ],
-      "purl": "pkg:npm/%40blamejs/exceptd-skills@0.13.25",
+      "purl": "pkg:npm/%40blamejs/exceptd-skills@0.13.27",
       "hashes": [
         {
           "alg": "SHA-256",
-          "content": "bf26fba4dd6ece0592fcccfd7e58fd670bc1b5b46f9f01bda086e42b3b4b30a4"
+          "content": "0e6a0230f25fee37c08637aad469a3dc640b544ff7b8b426ddeb8f59c83056d5"
         }
       ],
       "externalReferences": [
         {
           "type": "distribution",
-          "url": "https://www.npmjs.com/package/@blamejs/exceptd-skills/v/0.13.25"
+          "url": "https://www.npmjs.com/package/@blamejs/exceptd-skills/v/0.13.27"
         },
         {
           "type": "vcs",
@@ -116,11 +116,11 @@
       "hashes": [
         {
           "alg": "SHA-256",
-          "content": "fb8caf215ddd040d88701f1398d9f959e16a63a2fd7fac3392bdca6af2926a0b"
+          "content": "3b36f7f22962fdb7d595369940fc1cb5a4f58e22bb531da6dc56f58d1af7ef3f"
         },
         {
           "alg": "SHA3-512",
-          "content": "6fc54dda4e3cc3b1c5148ea2b66358c71d78b81b96cf844cdaa59bd8739abd76435e218a81957e2f3a30b990ff8ff8f0a7af6ba92d2d227584f2848c60ef579d"
+          "content": "c670a08d9db2838a25bab62aed25a06b47236cd3ca278d243ec951b143d2e98c9cf2c1eecc83966936d02921c1b00fa7837176d09ece03d629f78592545e7db5"
         }
       ]
     },
@@ -281,11 +281,11 @@
       "hashes": [
         {
           "alg": "SHA-256",
-          "content": "f375c1932e65be65b688db7069956dc97b3bcc3fe5492446fc3f5f3ea9c4b52e"
+          "content": "a8b22536e4d63b36a63018bf1024a407cc3533be7ee428e0b414ad6991f623ad"
         },
         {
           "alg": "SHA3-512",
-          "content": "ceb36659799f19868b5d32af014de1a315279c2321538d7aaa73a7222295c9b5db882a402de04c3c8299f54dfda6743aea9900d215ffd6f1ebcf9b2badd2661a"
+          "content": "d6511abcb934137785dac21dbd0a62a55e6e1e67a9c187cc834112db2ca82ed761bcd43321e5334069904dd0eda3b1d07b87ca24e8dfe490bfc75da5186747ba"
         }
       ]
     },
@@ -1031,11 +1031,11 @@
       "hashes": [
         {
           "alg": "SHA-256",
-          "content": "34b9a8ed6d468fe83422e7fa1aae30d205ccd01e8f9e9e3f4a2273ab10f5cfc5"
+          "content": "7d129a3edcb6ee6ff77622fb7ab2a9b9dd45a218d7a74d209edc72375c1d4cba"
         },
         {
           "alg": "SHA3-512",
-          "content": "fa5b647c27f0252d9ae1eae179a1b9e10c3186a87328f428c716c7adc74eac347aea5c6094fb7997295d32b3bfd18536709e1dba9659a4b807a274a233d227d8"
+          "content": "4411999d653bceaa72d15b1f9bfa267949a233e09161a003af49da39c8aaca533143428f6ea5d7fade114a744c28c804b1c7c2c5e58b993fa921ea1ae9ab3a48"
         }
       ]
     },
@@ -1451,11 +1451,11 @@
       "hashes": [
         {
           "alg": "SHA-256",
-          "content": "56140e0d1624f8c83333f27759e21815daf57a8face69d7aff3eb88d087f27b1"
+          "content": "9e9f77b63bb9cf8a34e235fe103e3e71efa558f2802fd563f73b18dc8f516d99"
         },
         {
           "alg": "SHA3-512",
-          "content": "2a89b19e1c07e14ea1f39f430cc08efe9a9da99373851014ade1c82192dce9c5bb67521ea1ff2985bfb1b731ce81ee50c032d5a314d02ebc89e21bc24c08052f"
+          "content": "5f7c38510b2a99fdbec98735d32ffb951ed2969578ab0d86a18c056df700d5fad264f4316d8311b0cc948797cf0e4c12cc6874e1fb13c4f25bdc01b376a3c1b2"
         }
       ]
     },
@@ -1939,6 +1939,21 @@
         }
       ]
     },
+    {
+      "bom-ref": "file:scripts/check-version-tags.js",
+      "type": "file",
+      "name": "scripts/check-version-tags.js",
+      "hashes": [
+        {
+          "alg": "SHA-256",
+          "content": "ab58a0d6a5f43e94f63161af45b4e436fe50f898e9a44588bdc15103dbc31aa8"
+        },
+        {
+          "alg": "SHA3-512",
+          "content": "36a39edf675786de2e7417e6097cd18f03f499ef994b8d728f93f1ca7b657e5c577f2b6ad7360a3470e7fba6287cbc65ae1eb98578c0d4ce689bd9e0f533c398"
+        }
+      ]
+    },
     {
       "bom-ref": "file:scripts/hooks/pre-commit.sh",
       "type": "file",
@@ -1961,11 +1976,11 @@
       "hashes": [
         {
           "alg": "SHA-256",
-          "content": "306e02cc802bc1ca6fe98fb1e4e57c895c7f458895cb2acc459dc6f7ba0127c9"
+          "content": "b1edfc22514b633f9932382d5d0fa091fa28b83535b4f0dac764ea2f4aebc85d"
         },
         {
           "alg": "SHA3-512",
-          "content": "e851348e5f7bec18a3a2a3757d1f5720cc0892c0bf527304e5e79dfc9931c7309ba6fd53918fcd63c2e7cd880487642af8f06d0b1195e0620e2aa9fadf643ba6"
+          "content": "f2b2d68424127b87e0f6a5a7cd2e64e4910d26af9ef4279f0aac49c81fdd9a47617013cde0165066a7678db1bee3a3a4c75882228befb94cad5219125dbb8fd9"
         }
       ]
     },

package/scripts/check-version-tags.js

@@ -0,0 +1,259 @@
+#!/usr/bin/env node
+"use strict";
+
+/**
+ * scripts/check-version-tags.js
+ *
+ * Refuses NEW version-stamped comments / filenames in the tracked
+ * source tree. The authoritative version surfaces are:
+ *
+ *   1. package.json / manifest.json `"version"` field
+ *   2. CHANGELOG.md `## X.Y.Z` headings
+ *   3. git tags
+ *   4. CLI `version` verb output (reads from package.json)
+ *
+ * Anywhere else, `// v0.13.22` / `Pre-v0.13.22` / `*-v0_13_22.test.js`
+ * is phase residue — operators don't have the roadmap, version tags
+ * rot the moment the next release lands, and `git clone` ships every
+ * comment to operators along with the code.
+ *
+ * The check uses a baseline snapshot (`tests/.version-tag-baseline.
+ * json`) capturing current violation counts per file. Future scans
+ * compare against the baseline:
+ *
+ *   - Filename violations beyond baseline → fail.
+ *   - Comment violations beyond baseline (in any file)        → fail.
+ *   - Violations strictly within baseline                     → ok.
+ *   - Violations below baseline (drift reduced)               → ok +
+ *     suggestion to refresh the baseline.
+ *
+ * Refresh: `node scripts/check-version-tags.js --update-baseline`.
+ *
+ * Wired into `npm run predeploy` as a gate.
+ */
+
+const fs = require("node:fs");
+const path = require("node:path");
+
+const ROOT = path.join(__dirname, "..");
+const BASELINE_PATH = path.join(ROOT, "tests", ".version-tag-baseline.json");
+
+// Directories we do not walk at all.
+const SKIP_DIRS = new Set([
+  "node_modules", ".git", ".keys", ".cache", ".scratch",
+  "data", "vendor", ".husky",
+]);
+
+// File extensions we scan for comment violations.
+const SCAN_EXTS = new Set([".js", ".cjs", ".mjs", ".md"]);
+
+// Paths that the project intentionally version-stamps:
+//   - CHANGELOG headings are how operators navigate the file
+//   - package.json / manifest.json carry the canonical version field
+//   - manifest-snapshot.json + sbom.cdx.json contain version-pinned
+//     metadata (the SBOM IS a version-stamped manifest)
+//   - lib/version-pins.js is a version-constant lookup table
+//   - This checker itself documents what it forbids
+//   - .git-blame-ignore-revs carries commit hashes, not version tags,
+//     but is conventional config the user maintains
+const COMMENT_EXEMPT = new Set([
+  "package.json",
+  "manifest.json",
+  "manifest-snapshot.json",
+  "sbom.cdx.json",
+  "CHANGELOG.md",
+  "lib/version-pins.js",
+  "scripts/check-version-tags.js",
+  // Gitignored local-only contributor docs — never shipped.
+  "CLAUDE.md",
+]);
+
+// Pattern: project version like `v0.13.22` or bare `0.13.22`. Matches
+// our pre-1.0 release range. External package versions like ATLAS
+// `v5.6.0` or CycloneDX `1.6` don't match because the major is 0.
+const VERSION_TAG_RE = /\bv?0\.\d+\.\d+\b/;
+
+// Phase residue patterns — broader than just version tags.
+const PHASE_RESIDUE_RES = [
+  /\bcycle\s+\d+\b/i,        // "cycle 13 P3 F3"
+  /\bphase\s+\d+(\.\d+)+\b/i,// "phase 9.11k"
+  /\bPre-v?0\.\d+\.\d+\b/i,  // "Pre-0.13.22"
+];
+
+// Filename pattern. `-v0_13_22.test.js` style.
+const FILENAME_VERSION_RE = /-v\d+_\d+_\d+\.test\.\w+$/;
+
+function walk(dir, results = []) {
+  for (const name of fs.readdirSync(dir)) {
+    if (SKIP_DIRS.has(name)) continue;
+    const full = path.join(dir, name);
+    const rel = path.relative(ROOT, full).replace(/\\/g, "/");
+    let stat;
+    try { stat = fs.statSync(full); }
+    catch { continue; }
+    if (stat.isDirectory()) {
+      walk(full, results);
+    } else if (stat.isFile()) {
+      results.push(rel);
+    }
+  }
+  return results;
+}
+
+function countCommentViolations(rel) {
+  if (COMMENT_EXEMPT.has(rel)) return 0;
+  const ext = path.extname(rel);
+  if (!SCAN_EXTS.has(ext)) return 0;
+  let text;
+  try { text = fs.readFileSync(path.join(ROOT, rel), "utf8"); }
+  catch { return 0; }
+  let count = 0;
+  for (const line of text.split(/\r?\n/)) {
+    if (VERSION_TAG_RE.test(line)) { count++; continue; }
+    for (const re of PHASE_RESIDUE_RES) {
+      if (re.test(line)) { count++; break; }
+    }
+  }
+  return count;
+}
+
+function scanCurrent() {
+  const files = walk(ROOT);
+  const byFile = {};
+  const filenameViolations = [];
+  for (const rel of files) {
+    if (FILENAME_VERSION_RE.test(rel)) filenameViolations.push(rel);
+    const n = countCommentViolations(rel);
+    if (n > 0) byFile[rel] = n;
+  }
+  return { byFile, filenameViolations };
+}
+
+function readBaseline() {
+  if (!fs.existsSync(BASELINE_PATH)) {
+    return { byFile: {}, filenameViolations: [], _missing: true };
+  }
+  try {
+    return JSON.parse(fs.readFileSync(BASELINE_PATH, "utf8"));
+  } catch (e) {
+    console.error(`[check-version-tags] baseline at ${path.relative(ROOT, BASELINE_PATH)} is malformed: ${e.message}`);
+    process.exitCode = 2;
+    return null;
+  }
+}
+
+function writeBaseline(current) {
+  const body = {
+    note: "Snapshot of pre-existing version-tag drift. The check-version-tags gate fails when these counts go UP. Refresh after an organic cleanup with `node scripts/check-version-tags.js --update-baseline`.",
+    recorded_at: new Date().toISOString().split("T")[0],
+    byFile: current.byFile,
+    filenameViolations: current.filenameViolations,
+  };
+  fs.writeFileSync(BASELINE_PATH, JSON.stringify(body, null, 2) + "\n");
+  console.log(`[check-version-tags] wrote baseline to ${path.relative(ROOT, BASELINE_PATH)}`);
+  console.log(`  ${Object.keys(current.byFile).length} file(s) with comment violations`);
+  console.log(`  ${current.filenameViolations.length} filename violation(s)`);
+}
+
+function main() {
+  const wantUpdate = process.argv.includes("--update-baseline");
+  const current = scanCurrent();
+
+  if (wantUpdate) {
+    writeBaseline(current);
+    process.exitCode = 0;
+    return;
+  }
+
+  const baseline = readBaseline();
+  if (!baseline) return;
+
+  if (baseline._missing) {
+    console.error(`[check-version-tags] baseline missing at ${path.relative(ROOT, BASELINE_PATH)}.`);
+    console.error("Run `node scripts/check-version-tags.js --update-baseline` to capture the current state.");
+    process.exitCode = 2;
+    return;
+  }
+
+  const regressions = [];
+
+  // Filename regressions: any new filename matching the pattern that
+  // wasn't in the baseline.
+  for (const rel of current.filenameViolations) {
+    if (!baseline.filenameViolations.includes(rel)) {
+      regressions.push({
+        kind: "filename",
+        path: rel,
+        reason: "new test filename carries a version tag — rename to describe the surface it pins, not the release",
+      });
+    }
+  }
+
+  // Comment regressions: per-file count grew.
+  for (const [rel, n] of Object.entries(current.byFile)) {
+    const prior = baseline.byFile[rel] || 0;
+    if (n > prior) {
+      regressions.push({
+        kind: "comment",
+        path: rel,
+        baseline: prior,
+        actual: n,
+        reason: `comment-level version-tag count grew from ${prior} to ${n} — describe the WHY of the current code, not the release that introduced it`,
+      });
+    }
+  }
+
+  // Files newly added to the violation set (not in baseline at all).
+  for (const rel of Object.keys(current.byFile)) {
+    if (!(rel in baseline.byFile)) {
+      const n = current.byFile[rel];
+      // Skip if already captured as a count regression above.
+      if (regressions.some(r => r.path === rel)) continue;
+      regressions.push({
+        kind: "comment",
+        path: rel,
+        baseline: 0,
+        actual: n,
+        reason: `new file carries ${n} version-tag comment(s) — describe the WHY of the current code, not the release that introduced it`,
+      });
+    }
+  }
+
+  if (regressions.length === 0) {
+    const totalFiles = Object.keys(current.byFile).length;
+    const totalFilenames = current.filenameViolations.length;
+    console.log(`[check-version-tags] ok — no new version tags. (${totalFiles} file(s) within baseline, ${totalFilenames} legacy filename(s).)`);
+    process.exitCode = 0;
+    return;
+  }
+
+  console.error(`[check-version-tags] FAIL — ${regressions.length} new version-tag regression(s).`);
+  console.error("");
+  for (const r of regressions) {
+    if (r.kind === "filename") {
+      console.error(`  ${r.path}`);
+      console.error(`    → ${r.reason}`);
+    } else {
+      console.error(`  ${r.path}  (${r.baseline} → ${r.actual})`);
+      console.error(`    → ${r.reason}`);
+    }
+  }
+  console.error("");
+  console.error("Authoritative version surfaces (mentions are LOAD-BEARING here):");
+  console.error("  package.json / manifest.json `version` field");
+  console.error("  CHANGELOG.md `## X.Y.Z` headings (body should describe behavior, not compare versions)");
+  console.error("  git tags");
+  console.error("");
+  console.error("Fix anywhere else:");
+  console.error("  - Rename test files to describe the surface (no `-v0_X_Y` suffix).");
+  console.error("  - Rewrite comments to describe the WHY of the current code, not the release.");
+  console.error("  - In CHANGELOG bodies, use 'Previously' / 'Now' phrasing, not 'Pre-X.Y.Z'.");
+  console.error("");
+  console.error("If a violation is legitimate (e.g. a deprecation timeline that needs a specific version), add the");
+  console.error("file path to COMMENT_EXEMPT in scripts/check-version-tags.js with a justifying comment.");
+
+  process.exitCode = 1;
+}
+
+if (require.main === module) main();
+

package/scripts/predeploy.js

@@ -207,6 +207,20 @@ const GATES = [
     args: [path.join(ROOT, "scripts", "check-catalog-gap-budget.js")],
     ciJobName: "Data integrity (catalog + manifest snapshot)",
   },
+  {
+    // Version-tag drift gate. Compares the tracked tree against a
+    // baseline snapshot of pre-existing `// vX.Y.Z` comments and
+    // `*-vX_Y_Z.test.js` filenames. Fails on NEW additions outside
+    // the authoritative version surfaces (package.json /
+    // manifest.json / CHANGELOG headings / git tags). The full rule
+    // is documented at the top of check-version-tags.js; refresh the
+    // baseline after an organic cleanup via
+    // `node scripts/check-version-tags.js --update-baseline`.
+    name: "Version-tag drift (no new phase residue)",
+    command: process.execPath,
+    args: [path.join(ROOT, "scripts", "check-version-tags.js")],
+    ciJobName: "Data integrity (catalog + manifest snapshot)",
+  },
 ];
 
 function runGate(gate) {