@openwop/openwop-conformance 2.46.10 → 2.46.12

package/CHANGELOG.md

@@ -1,3 +1,15 @@
+## [2.46.12] — 2026-10-16 — RFC 0245 and RFC 0247 Active
+
+- **No assertion changes yet.** RFC 0245 (eval fixtures replay a recorded tool error as an error) is `Active`: the packed v2 `agent-eval-suite` schema gains optional `toolResponses[].isError`, the capability schema gains `agents.evalSuite.fixtureErrors`, and three gap rows change the packed `spec/v1/gaps.json`. The fixture and the two gated legs follow.
+- **RFC 0247 `Active`: bounded stream lifetime.** The packed v2 surfaces gain the `resumable` query parameter on `streamRunEvents` and `streamHostEvents`, and the limits facet gains `maxStreamAgeSec`. No new leg (gap G1: a bounded close needs a long-held live run). A host that advertises nothing new, and a consumer that never sends `resumable=true`, see no change.
+
+## [2.46.11] — 2026-10-15 — a shared id reads the same on every run
+
+- **Ledger reader: one requirement id recorded by two scenario files.** Each file runs in its own worker, and `readLedgerFile` kept whichever line came last, so the row's `assertions` and `scenario` belonged to the worker that finished last. openwop-app's `prompts.resolution-precedes-composition` read 9 then 5 assertions on identical results.
+  - The least certifiable file still decides the result and keeps its detail, so a failure in either file is never masked (unchanged).
+  - New: the files' assertion counts sum, as the in-worker fold already sums legs, and a tie names the first file by name.
+  - **What a host's next bundle shows:** rows witnessed by two files carry the summed count and a fixed scenario. No result changes.
+
 ## [2.46.10] — 2026-10-14 — RFC 0244 Accepted (provisional)
 
 - **No assertion changes.** RFC 0244 flips to `Accepted` (provisional) on MyndHyve's certified 2.46.9 cut, so the packed `spec/v1/gaps.json` changes (G9, G10 and G11 close).

package/dist/lib/requirement-ledger.js

@@ -136,8 +136,12 @@ export function recordRequirement(requirementId, disposition, detail, extras) {
 export function readLedgerFile(path) {
     if (!existsSync(path))
         return [];
-    const rank = { 'executed-fail': 0, blocked: 1, 'executed-pass': 2, skipped: 3, inapplicable: 4 };
-    const merged = new Map();
+    const rank = CERTIFIABILITY_RANK;
+    // Pass 1, per (id, scenario file). `<=`, not `<`: a per-`it` FOLD (2.37.0)
+    // appends the cumulative row after the leg rows it folded, so on an equal
+    // disposition the LAST line is the one carrying the summed `assertionCount`.
+    // For genuinely duplicate lines the two are identical and the choice is a no-op.
+    const perFile = new Map();
     for (const line of readFileSync(path, 'utf8').split('\n')) {
         if (line.trim() === '')
             continue;
@@ -150,15 +154,32 @@ export function readLedgerFile(path) {
         }
         if (typeof e.requirementId !== 'string' || !DISPOSITIONS.includes(e.disposition))
             continue;
-        const prior = merged.get(e.requirementId);
-        // `<=`, not `<`: a per-`it` FOLD (2.37.0) appends the cumulative row after
-        // the leg rows it folded, so on an equal disposition the LAST line is the
-        // one carrying the summed `assertionCount`. For genuinely duplicate lines
-        // the two are identical and the choice is a no-op.
+        let files = perFile.get(e.requirementId);
+        if (files === undefined)
+            perFile.set(e.requirementId, (files = new Map()));
+        const file = e.scenarioFile ?? '';
+        const prior = files.get(file);
         if (prior === undefined || rank[e.disposition] <= rank[prior.disposition])
-            merged.set(e.requirementId, e);
+            files.set(file, e);
     }
-    return [...merged.values()].sort((a, b) => a.requirementId.localeCompare(b.requirementId));
+    // Pass 2, across files (2.46.11). Two files that witness ONE id run in
+    // different workers, so "last line wins" made the row's `assertionCount` and
+    // `scenarioFile` whichever worker finished last: the same host read 9 then 5
+    // assertions for `prompts.resolution-precedes-composition` with no result
+    // change. The least certifiable file still decides the disposition and keeps
+    // its detail; counts sum, as the in-worker fold sums them; and a tie names
+    // the first file by name, so the row is the same on every run.
+    const merged = [];
+    for (const files of perFile.values()) {
+        const rows = [...files.values()].sort((a, b) => rank[a.disposition] - rank[b.disposition] || (a.scenarioFile ?? '').localeCompare(b.scenarioFile ?? ''));
+        const winner = rows[0];
+        if (rows.length === 1 || !rows.some((r) => r.assertionCount !== undefined)) {
+            merged.push(winner);
+            continue;
+        }
+        merged.push({ ...winner, assertionCount: rows.reduce((n, r) => n + (r.assertionCount ?? 0), 0) });
+    }
+    return merged.sort((a, b) => a.requirementId.localeCompare(b.requirementId));
 }
 /**
  * The disposition for a requirement. **Absent resolves to `blocked`, never to a

package/dist/spec-artifacts.lock.json

@@ -1,5 +1,5 @@
 {
   "package": "@openwop/spec-artifacts",
-  "version": "2.46.10",
-  "stampSha256": "89832e47efa852d9a9528dd1528aacb7b7c7e803df7d905b6be3ab3c105ecbce"
+  "version": "2.46.12",
+  "stampSha256": "fec6d911af8113b65da78d98ef8c21f80ac8bb2d0a5f6303287dd5d88409cc1a"
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openwop/openwop-conformance",
-  "version": "2.46.10",
+  "version": "2.46.12",
   "description": "Black-box conformance suite for OpenWOP hosts (v2 current; v1 scenarios retained for the frozen v1 tree).",
   "repository": {
     "type": "git",
@@ -56,6 +56,6 @@
     "@openwop/spec-artifacts": "file:../spec-artifacts"
   },
   "peerDependencies": {
-    "@openwop/spec-artifacts": "2.46.10"
+    "@openwop/spec-artifacts": "2.46.12"
   }
 }

package/schemas/CORPUS-STAMP.json

@@ -1,17 +1,17 @@
 {
   "_comment": "Provenance of @openwop/spec-artifacts (RFC 0168 §D.2). files: SHA-256 per file; the conformance suite compares the installed peer against dist/spec-artifacts.lock.json at start.",
   "package": "@openwop/spec-artifacts",
-  "version": "2.46.10",
-  "corpusTag": "v2.46.10",
+  "version": "2.46.12",
+  "corpusTag": "v2.46.12",
   "files": {
     "api/.redocly.lint-ignore.yaml": "bf5a8350b88a72fa43f59605ed8d903ed24b6cfccda5e45509c9f6ed9ee4e712",
     "api/asyncapi.yaml": "d5ecb9ee6114582be3b1f662c84bfac9ae96dae7bacb853e461168f70a8e1c7d",
     "api/grpc/openwop.proto": "c3e72bb17cba514ee98feb6434e6c9b6ea6795bfd086489ec69fd882dd1ad977",
     "api/openapi.yaml": "ac6de5303872664b78ff419c035915593e01599e02835e777a0123de1baeb398",
     "api/redocly.yaml": "b0604c89b2ca6d5076ec25725c539dad44a741a811fe524439ee6daef8baa09f",
-    "api/seams-v2.yaml": "d432df92e9d6f9b5f507356bb055937dd5289e7b9f84b1c8ec64ac422bb36929",
-    "api/v2/asyncapi.yaml": "c7fb02dd90ca56745a8f1dc49f8808a74bc45fc5c7e6252f3724bb273904cf1b",
-    "api/v2/openapi.yaml": "780e4d7ba44543dd17647091837c1fa5978e658576c7c8ad6fcc105425a51b24",
+    "api/seams-v2.yaml": "e70dca4462c179fe7bfcc25244e481aa04c3da2a97e74d99e70c562191b9ec0b",
+    "api/v2/asyncapi.yaml": "964c60cadd09209e704b52fce5f68c28f7ea08e8bd9c04318c931e1969d79d64",
+    "api/v2/openapi.yaml": "0aa511e8f9c6b49e224b111a2afe81ef48fc92b1fc77dd6287b04f2e6597ea33",
     "api/v2/redocly.yaml": "1e66b60e6118ad11a823bb620678be464d99dfe50a40e3e6f93ec9429b88b34c",
     "schemas/README.md": "276fc1d51eb08e14e9127b9fd27761d7046ee33c12000f3a64a801322db8bf58",
     "schemas/a2a-task-state.schema.json": "75d5049dea8bd873ff0e7546f1c60c8d36c219bec7084264360be54a8be30ae1",
@@ -103,7 +103,7 @@
     "schemas/v2/a2ui-surface-delta-frame.schema.json": "6889c5da3fe7e76cb614038c0eea6774776fa640448558b258eb87fb50cb9b64",
     "schemas/v2/agent-deployment-transition.schema.json": "93a334ff042d51378c89fdd7134664271bf9592419bb5e3e00821ed06bc1904e",
     "schemas/v2/agent-deployment.schema.json": "d59b43fb62b522005ee22bc3bc53775a6777493e5e42d71812abc347625fa7fe",
-    "schemas/v2/agent-eval-suite.schema.json": "e80439bc6873a5d9a14122f81d16f7650104018d814bd316aafd3fc8b8225604",
+    "schemas/v2/agent-eval-suite.schema.json": "8b6dac5c6485c1fc4b7228efc0a87fa5c64fce89ddd228d4f92786b1b63e37b3",
     "schemas/v2/agent-inventory-response.schema.json": "5f74c3da946dfc3114624567a989717748081ab947b4d2af3f49034d67f4306b",
     "schemas/v2/agent-manifest.schema.json": "004ff150e955deebb9e64273de5093f5144b9b3895b09953e65aa74b61140778",
     "schemas/v2/agent-org-chart.schema.json": "113b079e1dce95908467cbef7f83e076fdb82aab5b0678fc1a43e32bfc9cbadf",
@@ -117,7 +117,7 @@
     "schemas/v2/artifact.schema.json": "85b9dc284d39b99d2a66fa8cbf47f2dd3d8526d3bd6cc411ad315b372de38975",
     "schemas/v2/audit-verify-result.schema.json": "14af8e9283510f0a40e48a2059c929ee44e0fc9fa3cc66a95bfbe90b17c48297",
     "schemas/v2/budget-policy.schema.json": "1de68095db0beb5c7a4be629de26e79f62ebbe748a88d46f36e768edfd1f0d68",
-    "schemas/v2/capabilities.schema.json": "a717d34d98201cefe1769421c468d22e124d1a46881f60683cb6e103843dbc9a",
+    "schemas/v2/capabilities.schema.json": "87faea080e8043abd16cba19dece777a810731fc3e4455c96c02d0bfd0997d84",
     "schemas/v2/certification-bundle.schema.json": "5cfa7fb5247160333e33a58b60d42bcd4e5269d8391d6a112d301e14828892ed",
     "schemas/v2/channel-presence-payload.schema.json": "fc03df0c20cc12a25f7ee6a47503a4c5931d79fe3d5ff24815e8fd7c1bcb9a87",
     "schemas/v2/channel-written-payload.schema.json": "ccecff3c71a3275ad8db035ac0e04db25e6f8d23cf091ddbcd5a60d1be1cbb6b",
@@ -208,14 +208,14 @@
     "schemas/workspace-file.schema.json": "464de85c2a068243084ee9c1d969bc7cd5d8f7948574e58450d6493c38a0e1e4",
     "spec/v1/alias-detectors.json": "22444c22bb88c568339ef512553f9fc38146e22b566d11ab8608f655d688d19c",
     "spec/v1/capability-declaration-classes.json": "e7729aed5c4b4e1dd02abab0530f14cc95f5d4070fe51fb139e7f5cccefa00c6",
-    "spec/v1/core-standard-manifest.json": "ddfa3fb570b38ffa11ed013fcf56ee6ce6639f80030460a4589dc7de105da292",
+    "spec/v1/core-standard-manifest.json": "920cf7ccccfb66f6438fc72a477166d3757c5c6ca6e26d7d247f0a3b801334a0",
     "spec/v1/deprecations.json": "e723bc11d7272c71789ad1303bb7f21809a8a024abf57594f9d5c35dc199ece9",
     "spec/v1/deprecations.schema.json": "3e393c405d2a41b467d8c5e3c468549078df1ce6a6d2588fc488097b95d9b55b",
     "spec/v1/end-of-support.json": "35f0bf3b70fe78b16e2a2fd81a153fc27e703f6c16376c6fe7cdf1f4f2c4e6d4",
     "spec/v1/event-codemap.json": "3da60d884157793a360da532a9fcbbfb5285636db325a74cec94b34622186d97",
     "spec/v1/event-codemap.schema.json": "d05933b2e88103aff51a2f774df97b9bda065e0fc88dbd9b114cc47a32189174",
     "spec/v1/extensions.json": "79a60754aa16cbdbdb604f8e5c00038af535a1690a4d07bd5ff7c38f0256a7bc",
-    "spec/v1/gaps.json": "2fc8bb0c3196c62239a10dca000901b77d260b97776893f117eb5d8d144042a9",
+    "spec/v1/gaps.json": "07c606b9abbcefbb5fcc18a230765db695ceba9d98f3e7da9dd92b8caebbfd7f",
     "spec/v1/gaps.schema.json": "8fd83259f556553c9df0f53e7a82ca8c2a4771d471197f69b4e2ae0ceeacfd66",
     "spec/v1/migrations.json": "4c50760be70cb644d02427bcaf5749606700edf4513f423903cf3b1d12f5c300",
     "spec/v1/migrations.schema.json": "886779aa6c22e646db097f5df210adb018a4dd14a7b815465a18c8a7056c8f72",
@@ -228,7 +228,7 @@
     "spec/v2/core/connection-packs.md": "c80f95bd6a03698f35c72cdb2b46bd973f49203bbd786c26760e35fa86e6f23d",
     "spec/v2/core/conversation.md": "8d6abedf6af408a74577753bfca24719d318e2632e10b45ff5d63bdcf31a8aaa",
     "spec/v2/core/errors.md": "e48838701b86bbfe0ffee77b18a8ffc4e5f04724b4318507d9771da653f8da0f",
-    "spec/v2/core/events.md": "370621629c2439747f2e4ddab4780eea4345354f79e5098739d66823873f3105",
+    "spec/v2/core/events.md": "c3dcece6f1b1aa43e1e01c5ece23b2df075a505d5f6c2f37de0801aeaa486b62",
     "spec/v2/core/execution.md": "4218dbb397534940f83ca5d46561f02bf8f3fdf68d2ae7fefe7f2eb2ed771103",
     "spec/v2/core/form-content-packs.md": "02931227899676fd4581a27017be2a6c4953904717be9b81c29d12f87ab3917d",
     "spec/v2/core/headers.md": "f69fe81e4a3ad21879bb85948660c16860eb05bf152928c31aa4898f6d7d1887",
@@ -254,7 +254,7 @@
     "spec/v2/core/workflow-chain-packs.md": "fdce7a8477e269950a2bac05ee6eddbd66eb0b670b464d00054f7196ec05d29c",
     "spec/v2/corrections.json": "3a8d93e33865b692d26a3b6bfd3819766e206b2e0c2a69669baa4bf833d05a49",
     "spec/v2/corrections.schema.json": "46bbffb954d13bd9995aebc912ceeb2544f7064331abf25fd6ab85c81a5cec94",
-    "spec/v2/declaration.json": "773af65df59ce7bc2b3ab17093e23ac8083a8faaeff643302fe30198deae933b",
+    "spec/v2/declaration.json": "de8558d76a03d07101c5e26a7311d3e90b449f065c9ba418f4abd9c91864a58b",
     "spec/v2/declaration.schema.json": "ef951633e4070899f64828f5cf1b5f4870270a0360f81dbb49b5b4d6cd407550",
     "spec/v2/eos-override.json": "4c72d2f6a658943b01c1de1643cf294642091933af7832f38ee5495df684cb2d",
     "spec/v2/errors.json": "1f0e29455edbd22e28f67f221766209c3c55ac4987beb77993b5702dd7cae131",
@@ -281,7 +281,7 @@
     "spec/v2/ext/sandbox-runtime-notes/README.md": "0d1290fb1e20b2e35c0b15051541ab6546e4e2a6f693b4f912cb3f7d6a65e92c",
     "spec/v2/ext/webResearch/README.md": "b675e6f6646128d781b661f754c2393eed86e5935d16c3a22df582a5ce025b9b",
     "spec/v2/facets/a2a.schema.json": "b68bf5e29e455e46c2ad7cfcccf5963730ec2310ff9db81884550d8df30fd6ef",
-    "spec/v2/facets/agents.schema.json": "62a979d23219ca684bc16f91f81a4c92c2cb56e0f084d2e4f694c03e0c7125df",
+    "spec/v2/facets/agents.schema.json": "88fe144572c25e1b09b99fd83e2ad12b2611d4d7366b5d53c3a50d0310faad93",
     "spec/v2/facets/aiProviders.schema.json": "4eb911c2403b8f618325194855a1326680fbda2549dfbfe86a70dd2743f8fb12",
     "spec/v2/facets/auditLogIntegrity.schema.json": "9a7f81320c8f940c6f4129bc7dd895a4063231c5458a717a08c2cbf9bb886c46",
     "spec/v2/facets/auth.schema.json": "7d98bc81a0f7d307460d039eebfebd21388091bfe1d538762583940f94c0bd2a",
@@ -290,6 +290,7 @@
     "spec/v2/facets/heartbeat.schema.json": "81d7fa1895a8b0241bd07cf00a7e7905bef0fd14eb3427325261c62eee5746a6",
     "spec/v2/facets/hostEvents.schema.json": "973ff452cb91356b4c625d0b06c8d2ef335e18183de09e1afde8acead22290b3",
     "spec/v2/facets/interrupt.schema.json": "2110ef32c3f880bf706475ef4cf8ddf97ff06871b8e6003a5edbcf4439f1fb70",
+    "spec/v2/facets/limits.schema.json": "218fcbd34dd2cdf7548c29c8c76e1a6aea2b7d7df1ba32fbe201bd40917bc66f",
     "spec/v2/facets/mcp.schema.json": "7060657dfcff170f079beed160e3db7c1d3cc4410cd3cf64863baf6459f781cd",
     "spec/v2/facets/nondeterminismPolicy.schema.json": "e9dfaa32e0778a1a837fdad3b8a0673fccda9579ab08c8bfe9777362dd15e989",
     "spec/v2/facets/oauth.schema.json": "3a0cbe249d3bd5a5982349ae7ba2534776cc958e4a57e0fbeabf2bdab6e99e69",
@@ -313,9 +314,9 @@
     "spec/v2/path-manifest.json": "7b5679a4dad98c80d855361d88a98c8a4e045af15e310862a05b285b815dc9fc",
     "spec/v2/peer-dependency-aliases.json": "d10299280abee08258502925bc327293ee413e0108cd6e6ec75ff6110653308d",
     "spec/v2/profiles.json": "b382b261c7844798575f7c1b37c934f0da19846016a2337d883d4f5ef6e05479",
-    "spec/v2/release.json": "481c91e5b8cc4965a426e8d946f11c8831192f5c02b3d01295d89a98ff0b0e2a",
+    "spec/v2/release.json": "d20dc7e5d37f395199e68fe471244cff4f744790a010a4b0eb3c6c28ba4c26e4",
     "spec/v2/retention-floors.json": "eaf3722d95c79947af1d4269ef85117e126518c588cfcf1a2b21b97269f51624",
-    "spec/v2/surface-baseline.json": "f774b4481aea140f0b9f5b83474ab216763d5fa089e4c7567c830b84a163b4d8"
+    "spec/v2/surface-baseline.json": "54bd7632dc314ffbb180b99a35f1367ab9eca60471aae5b98c6a020923c47203"
   },
-  "corpusCommit": "b5d6f3aa9cdc0a1f4f3fd705c57842b8c1443ff2"
+  "corpusCommit": "ca6b4ad06520858154f4e3dd9ba146496f038e43"
 }

package/src/lib/requirement-ledger.ts

@@ -221,8 +221,12 @@ export function recordRequirement(
  */
 export function readLedgerFile(path: string): readonly LedgerEntry[] {
   if (!existsSync(path)) return [];
-  const rank: Record<Disposition, number> = { 'executed-fail': 0, blocked: 1, 'executed-pass': 2, skipped: 3, inapplicable: 4 };
-  const merged = new Map<string, LedgerEntry>();
+  const rank = CERTIFIABILITY_RANK;
+  // Pass 1, per (id, scenario file). `<=`, not `<`: a per-`it` FOLD (2.37.0)
+  // appends the cumulative row after the leg rows it folded, so on an equal
+  // disposition the LAST line is the one carrying the summed `assertionCount`.
+  // For genuinely duplicate lines the two are identical and the choice is a no-op.
+  const perFile = new Map<string, Map<string, LedgerEntry>>();
   for (const line of readFileSync(path, 'utf8').split('\n')) {
     if (line.trim() === '') continue;
     let e: LedgerEntry;
@@ -232,14 +236,32 @@ export function readLedgerFile(path: string): readonly LedgerEntry[] {
       continue;
     }
     if (typeof e.requirementId !== 'string' || !DISPOSITIONS.includes(e.disposition)) continue;
-    const prior = merged.get(e.requirementId);
-    // `<=`, not `<`: a per-`it` FOLD (2.37.0) appends the cumulative row after
-    // the leg rows it folded, so on an equal disposition the LAST line is the
-    // one carrying the summed `assertionCount`. For genuinely duplicate lines
-    // the two are identical and the choice is a no-op.
-    if (prior === undefined || rank[e.disposition] <= rank[prior.disposition]) merged.set(e.requirementId, e);
+    let files = perFile.get(e.requirementId);
+    if (files === undefined) perFile.set(e.requirementId, (files = new Map()));
+    const file = e.scenarioFile ?? '';
+    const prior = files.get(file);
+    if (prior === undefined || rank[e.disposition] <= rank[prior.disposition]) files.set(file, e);
   }
-  return [...merged.values()].sort((a, b) => a.requirementId.localeCompare(b.requirementId));
+  // Pass 2, across files (2.46.11). Two files that witness ONE id run in
+  // different workers, so "last line wins" made the row's `assertionCount` and
+  // `scenarioFile` whichever worker finished last: the same host read 9 then 5
+  // assertions for `prompts.resolution-precedes-composition` with no result
+  // change. The least certifiable file still decides the disposition and keeps
+  // its detail; counts sum, as the in-worker fold sums them; and a tie names
+  // the first file by name, so the row is the same on every run.
+  const merged: LedgerEntry[] = [];
+  for (const files of perFile.values()) {
+    const rows = [...files.values()].sort(
+      (a, b) => rank[a.disposition] - rank[b.disposition] || (a.scenarioFile ?? '').localeCompare(b.scenarioFile ?? ''),
+    );
+    const winner = rows[0]!;
+    if (rows.length === 1 || !rows.some((r) => r.assertionCount !== undefined)) {
+      merged.push(winner);
+      continue;
+    }
+    merged.push({ ...winner, assertionCount: rows.reduce((n, r) => n + (r.assertionCount ?? 0), 0) });
+  }
+  return merged.sort((a, b) => a.requirementId.localeCompare(b.requirementId));
 }
 
 /**