audrey 1.0.1 → 1.0.3

package/docs/paper/output/submission-bundle/benchmarks/output/leaderboard/guardbench-leaderboard.json

@@ -1,7 +1,7 @@
 {
   "schemaVersion": "1.0.0",
   "suite": "GuardBench leaderboard",
-  "generatedAt": "2026-05-15T17:52:13.304Z",
+  "generatedAt": "2026-05-29T13:33:23.840Z",
   "ranking": [
     "verified bundle",
     "adapter conformance",
@@ -28,9 +28,9 @@
         "evidenceRecall": 1,
         "redactionLeaks": 0,
         "latency": {
-          "p50Ms": 2.465,
-          "p95Ms": 30.791,
-          "maxMs": 30.791
+          "p50Ms": 3.09,
+          "p95Ms": 28.181,
+          "maxMs": 28.181
         }
       },
       "conformance": {
@@ -41,7 +41,7 @@
       },
       "source": {
         "dir": "benchmarks/output/submission-bundle",
-        "manifestGeneratedAt": "2026-05-15T17:52:13.050Z",
+        "manifestGeneratedAt": "2026-05-29T13:33:23.534Z",
         "fileCount": 17
       },
       "verification": {

package/docs/paper/output/submission-bundle/benchmarks/output/leaderboard/guardbench-leaderboard.md

@@ -1,7 +1,7 @@
 # GuardBench Leaderboard
 
-Generated: 2026-05-15T17:52:13.304Z
+Generated: 2026-05-29T13:33:23.840Z
 
 | Rank | Subject | Verified | Conformant | Full Contract | Decision Accuracy | Evidence Recall | Redaction Leaks | p95 Latency | Bundle |
 |---:|---|---:|---:|---:|---:|---:|---:|---:|---|
-| 1 | Audrey Guard | yes | yes | 100.0% | 100.0% | 100.0% | 0 | 30.791ms | benchmarks/output/submission-bundle |
+| 1 | Audrey Guard | yes | yes | 100.0% | 100.0% | 100.0% | 0 | 28.181ms | benchmarks/output/submission-bundle |

package/docs/paper/output/submission-bundle/benchmarks/output/submission-bundle/submission-manifest.json

@@ -1,7 +1,7 @@
 {
   "schemaVersion": "1.0.0",
   "suite": "GuardBench submission bundle",
-  "generatedAt": "2026-05-15T17:52:13.050Z",
+  "generatedAt": "2026-05-29T13:33:23.534Z",
   "sourceDir": "benchmarks/output",
   "subject": {
     "name": "Audrey Guard",
@@ -16,9 +16,9 @@
     "evidenceRecall": 1,
     "redactionLeaks": 0,
     "latency": {
-      "p50Ms": 2.465,
-      "p95Ms": 30.791,
-      "maxMs": 30.791
+      "p50Ms": 3.09,
+      "p95Ms": 28.181,
+      "maxMs": 28.181
     }
   },
   "conformance": {
@@ -44,8 +44,8 @@
   "files": [
     {
       "path": "guardbench-conformance-card.json",
-      "bytes": 1735,
-      "sha256": "65d4a10afe10178e4e4e3d508e0e06ce5a4b28713debc27cf4d9b60a46050d2f"
+      "bytes": 1748,
+      "sha256": "88cba808b893cde3530f4e29317d4cf9ae3b20f08456c58d97d4e5300f5f9298"
     },
     {
       "path": "guardbench-manifest.json",
@@ -54,13 +54,13 @@
     },
     {
       "path": "guardbench-raw.json",
-      "bytes": 43455,
-      "sha256": "3b78d1a2432e7d72752f96d9ac4b2b49cf6f59eb65548fbadb21ea6adbb86b37"
+      "bytes": 43466,
+      "sha256": "66d4b69087258638f3572a40e1fd59bb84067034f899eaa2c27eed2dde554b2b"
     },
     {
       "path": "guardbench-summary.json",
-      "bytes": 69350,
-      "sha256": "21023f230b761f1b43f8ecabe519dd6b320c62ad56f0b6aa28bbcf7a2c8838f5"
+      "bytes": 69359,
+      "sha256": "91f264dd889e2c639a6fc6d1b867bc228b94c84ed5120345e23dddb79c11ee74"
     },
     {
       "path": "schemas/guardbench-adapter-registry.schema.json",
@@ -125,7 +125,7 @@
     {
       "path": "validation-report.json",
       "bytes": 739,
-      "sha256": "570d28760ad3611ccf4f08e98281a3fe659730463d13e96ff31073c7ee10ce65"
+      "sha256": "284c40f608508d59e8cdc8047371dfcf0a9824a891ff37b77035241ab5de275f"
     }
   ]
 }

package/docs/paper/output/submission-bundle/benchmarks/output/submission-bundle/validation-report.json

@@ -1,5 +1,5 @@
 {
-  "generatedAt": "2026-05-15T17:52:13.050Z",
+  "generatedAt": "2026-05-29T13:33:23.534Z",
   "sourceValidation": {
     "ok": true,
     "dir": "benchmarks/output",

package/docs/paper/output/submission-bundle/benchmarks/output/summary.json

@@ -1,5 +1,5 @@
 {
-  "generatedAt": "2026-05-15T17:52:00.842Z",
+  "generatedAt": "2026-05-29T13:33:19.790Z",
   "command": "node benchmarks/run.js --provider mock --dimensions 64",
   "config": {
     "provider": "mock",
@@ -23,13 +23,13 @@
         "system": "Audrey",
         "scorePercent": 100,
         "passRate": 100,
-        "avgDurationMs": 93.58333333333333
+        "avgDurationMs": 14.75
       },
       {
         "system": "Vector Only",
         "scorePercent": 41.66666666666667,
         "passRate": 25,
-        "avgDurationMs": 0.25
+        "avgDurationMs": 0.4166666666666667
       },
       {
         "system": "Keyword + Recency",
@@ -41,7 +41,7 @@
         "system": "Recent Window",
         "scorePercent": 37.5,
         "passRate": 25,
-        "avgDurationMs": 0
+        "avgDurationMs": 0.08333333333333333
       }
     ],
     "overall_scope": "comparable_suites",
@@ -170,25 +170,25 @@
             "system": "Audrey",
             "scorePercent": 100,
             "passRate": 100,
-            "avgDurationMs": 127.125
+            "avgDurationMs": 14.75
           },
           {
             "system": "Vector Only",
             "scorePercent": 56.25,
             "passRate": 37.5,
-            "avgDurationMs": 0.25
+            "avgDurationMs": 0.5
           },
           {
             "system": "Keyword + Recency",
             "scorePercent": 50,
             "passRate": 37.5,
-            "avgDurationMs": 0.75
+            "avgDurationMs": 0.875
           },
           {
             "system": "Recent Window",
             "scorePercent": 50,
             "passRate": 37.5,
-            "avgDurationMs": 0
+            "avgDurationMs": 0.125
           }
         ],
         "byFamily": [
@@ -277,7 +277,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 564,
+                "durationMs": 22,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -289,7 +289,7 @@
               },
               {
                 "system": "Vector Only",
-                "durationMs": 0,
+                "durationMs": 1,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -301,7 +301,7 @@
               },
               {
                 "system": "Keyword + Recency",
-                "durationMs": 6,
+                "durationMs": 7,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -336,7 +336,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 19,
+                "durationMs": 13,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -371,7 +371,7 @@
               },
               {
                 "system": "Recent Window",
-                "durationMs": 0,
+                "durationMs": 1,
                 "passed": false,
                 "score": 0.5,
                 "summary": "retrieved expected evidence, but conflicting evidence still appeared later",
@@ -394,7 +394,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 287,
+                "durationMs": 13,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -457,7 +457,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 38,
+                "durationMs": 13,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -518,7 +518,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 26,
+                "durationMs": 13,
                 "passed": true,
                 "score": 1,
                 "summary": "correct abstention",
@@ -527,7 +527,7 @@
               },
               {
                 "system": "Vector Only",
-                "durationMs": 0,
+                "durationMs": 1,
                 "passed": false,
                 "score": 0.5,
                 "summary": "no leak, but retrieved tangential context",
@@ -574,7 +574,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 20,
+                "durationMs": 13,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -632,7 +632,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 22,
+                "durationMs": 19,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -693,7 +693,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 41,
+                "durationMs": 12,
                 "passed": true,
                 "score": 1,
                 "summary": "correct abstention",
@@ -750,13 +750,13 @@
             "system": "Audrey",
             "scorePercent": 100,
             "passRate": 100,
-            "avgDurationMs": 26.5
+            "avgDurationMs": 14.75
           },
           {
             "system": "Keyword + Recency",
             "scorePercent": 25,
             "passRate": 0,
-            "avgDurationMs": 0.25
+            "avgDurationMs": 0
           },
           {
             "system": "Vector Only",
@@ -821,7 +821,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 44,
+                "durationMs": 14,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -832,7 +832,7 @@
               },
               {
                 "system": "Vector Only",
-                "durationMs": 1,
+                "durationMs": 0,
                 "passed": false,
                 "score": 0.5,
                 "summary": "retrieved expected evidence, but conflicting evidence still appeared later",
@@ -879,7 +879,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 24,
+                "durationMs": 14,
                 "passed": true,
                 "score": 1,
                 "summary": "correct abstention",
@@ -899,7 +899,7 @@
               },
               {
                 "system": "Keyword + Recency",
-                "durationMs": 1,
+                "durationMs": 0,
                 "passed": false,
                 "score": 0.5,
                 "summary": "no leak, but retrieved tangential context",
@@ -932,7 +932,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 17,
+                "durationMs": 16,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -993,7 +993,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 21,
+                "durationMs": 15,
                 "passed": true,
                 "score": 1,
                 "summary": "retrieved expected evidence",
@@ -1004,7 +1004,7 @@
               },
               {
                 "system": "Vector Only",
-                "durationMs": 0,
+                "durationMs": 1,
                 "passed": false,
                 "score": 0,
                 "summary": "missed target evidence",
@@ -1055,7 +1055,7 @@
             "system": "Audrey",
             "scorePercent": 100,
             "passRate": 100,
-            "avgDurationMs": 58.75
+            "avgDurationMs": 14
           },
           {
             "system": "Vector Only",
@@ -1073,7 +1073,7 @@
             "system": "Recent Window",
             "scorePercent": 0,
             "passRate": 0,
-            "avgDurationMs": 0
+            "avgDurationMs": 0.25
           }
         ],
         "byFamily": [
@@ -1118,7 +1118,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 71,
+                "durationMs": 18,
                 "passed": true,
                 "score": 1,
                 "summary": "matched all required signals",
@@ -1176,7 +1176,7 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 18,
+                "durationMs": 13,
                 "passed": true,
                 "score": 1,
                 "summary": "matched all required signals",
@@ -1211,7 +1211,7 @@
               },
               {
                 "system": "Recent Window",
-                "durationMs": 0,
+                "durationMs": 1,
                 "passed": false,
                 "score": 0,
                 "summary": "missed required signals: decision:block, warning:must_follow, reflex:block",
@@ -1234,12 +1234,12 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 12,
+                "durationMs": 13,
                 "passed": true,
                 "score": 1,
                 "summary": "matched all required signals",
                 "topResults": [
-                  "guard_hardened:replay_rejected error:guard receipt already has an outcome: 01KRPC8H5YR3DZPZYEJCWPJQYP",
+                  "guard_hardened:replay_rejected error:guard receipt already has an outcome: 01KSSZ0Y0W5BPW0QA26ZDFWXGG",
                   "decision:caution verdict:caution risk:0.55 Caution: 1 memory signal, 1 medium severity found before acting.",
                   "warning:recent_failure severity:medium npm test failed 1x recently: Vitest failed with spawn EPERM Before re-running npm test, check what changed since the last failure."
                 ],
@@ -1292,12 +1292,12 @@
             "results": [
               {
                 "system": "Audrey",
-                "durationMs": 134,
+                "durationMs": 12,
                 "passed": true,
                 "score": 1,
                 "summary": "matched all required signals",
                 "topResults": [
-                  "guard_hardened:non_guard_receipt_rejected error:not a guard receipt: 01KRPC8HA0WVXFATHHNX8VBRYF",
+                  "guard_hardened:non_guard_receipt_rejected error:not a guard receipt: 01KSSZ0Y19FPR6G897395CDBTH",
                   "decision:go verdict:clear risk:0 No relevant memory risks, prior failures, or must-follow procedures were found."
                 ],
                 "retrievalSummary": "guard_hardened:non_guard_receipt_rejected error:not a guard receipt: 01K | decision:go verdict:clear risk:0 No relevant memory risks, prior failure"
@@ -1352,7 +1352,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 564,
+            "durationMs": 22,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1364,7 +1364,7 @@
           },
           {
             "system": "Vector Only",
-            "durationMs": 0,
+            "durationMs": 1,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1376,7 +1376,7 @@
           },
           {
             "system": "Keyword + Recency",
-            "durationMs": 6,
+            "durationMs": 7,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1411,7 +1411,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 19,
+            "durationMs": 13,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1446,7 +1446,7 @@
           },
           {
             "system": "Recent Window",
-            "durationMs": 0,
+            "durationMs": 1,
             "passed": false,
             "score": 0.5,
             "summary": "retrieved expected evidence, but conflicting evidence still appeared later",
@@ -1469,7 +1469,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 287,
+            "durationMs": 13,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1532,7 +1532,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 38,
+            "durationMs": 13,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1593,7 +1593,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 26,
+            "durationMs": 13,
             "passed": true,
             "score": 1,
             "summary": "correct abstention",
@@ -1602,7 +1602,7 @@
           },
           {
             "system": "Vector Only",
-            "durationMs": 0,
+            "durationMs": 1,
             "passed": false,
             "score": 0.5,
             "summary": "no leak, but retrieved tangential context",
@@ -1649,7 +1649,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 20,
+            "durationMs": 13,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1707,7 +1707,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 22,
+            "durationMs": 19,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1768,7 +1768,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 41,
+            "durationMs": 12,
             "passed": true,
             "score": 1,
             "summary": "correct abstention",
@@ -1824,7 +1824,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 44,
+            "durationMs": 14,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1835,7 +1835,7 @@
           },
           {
             "system": "Vector Only",
-            "durationMs": 1,
+            "durationMs": 0,
             "passed": false,
             "score": 0.5,
             "summary": "retrieved expected evidence, but conflicting evidence still appeared later",
@@ -1882,7 +1882,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 24,
+            "durationMs": 14,
             "passed": true,
             "score": 1,
             "summary": "correct abstention",
@@ -1902,7 +1902,7 @@
           },
           {
             "system": "Keyword + Recency",
-            "durationMs": 1,
+            "durationMs": 0,
             "passed": false,
             "score": 0.5,
             "summary": "no leak, but retrieved tangential context",
@@ -1935,7 +1935,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 17,
+            "durationMs": 16,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -1996,7 +1996,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 21,
+            "durationMs": 15,
             "passed": true,
             "score": 1,
             "summary": "retrieved expected evidence",
@@ -2007,7 +2007,7 @@
           },
           {
             "system": "Vector Only",
-            "durationMs": 0,
+            "durationMs": 1,
             "passed": false,
             "score": 0,
             "summary": "missed target evidence",
@@ -2058,7 +2058,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 71,
+            "durationMs": 18,
             "passed": true,
             "score": 1,
             "summary": "matched all required signals",
@@ -2116,7 +2116,7 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 18,
+            "durationMs": 13,
             "passed": true,
             "score": 1,
             "summary": "matched all required signals",
@@ -2151,7 +2151,7 @@
           },
           {
             "system": "Recent Window",
-            "durationMs": 0,
+            "durationMs": 1,
             "passed": false,
             "score": 0,
             "summary": "missed required signals: decision:block, warning:must_follow, reflex:block",
@@ -2174,12 +2174,12 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 12,
+            "durationMs": 13,
             "passed": true,
             "score": 1,
             "summary": "matched all required signals",
             "topResults": [
-              "guard_hardened:replay_rejected error:guard receipt already has an outcome: 01KRPC8H5YR3DZPZYEJCWPJQYP",
+              "guard_hardened:replay_rejected error:guard receipt already has an outcome: 01KSSZ0Y0W5BPW0QA26ZDFWXGG",
               "decision:caution verdict:caution risk:0.55 Caution: 1 memory signal, 1 medium severity found before acting.",
               "warning:recent_failure severity:medium npm test failed 1x recently: Vitest failed with spawn EPERM Before re-running npm test, check what changed since the last failure."
             ],
@@ -2232,12 +2232,12 @@
         "results": [
           {
             "system": "Audrey",
-            "durationMs": 134,
+            "durationMs": 12,
             "passed": true,
             "score": 1,
             "summary": "matched all required signals",
             "topResults": [
-              "guard_hardened:non_guard_receipt_rejected error:not a guard receipt: 01KRPC8HA0WVXFATHHNX8VBRYF",
+              "guard_hardened:non_guard_receipt_rejected error:not a guard receipt: 01KSSZ0Y19FPR6G897395CDBTH",
               "decision:go verdict:clear risk:0 No relevant memory risks, prior failures, or must-follow procedures were found."
             ],
             "retrievalSummary": "guard_hardened:non_guard_receipt_rejected error:not a guard receipt: 01K | decision:go verdict:clear risk:0 No relevant memory risks, prior failure"

package/docs/paper/output/submission-bundle/docs/paper/07-evaluation.md

@@ -24,14 +24,14 @@ These numbers measure Audrey's local call path under an in-process mock embeddin
 
 ## Behavioral Regression Result
 
-The current `benchmarks/output/summary.json` was generated on 2026-05-15T17:52:00.842Z with command `node benchmarks/run.js --provider mock --dimensions 64` (Ledger: E24). It reports:
+The current `benchmarks/output/summary.json` was generated on 2026-05-29T13:33:19.790Z with command `node benchmarks/run.js --provider mock --dimensions 64` (Ledger: E24). It reports:
 
 | System | Score Percent | Pass Rate | Average Duration Ms |
 |---|---:|---:|---:|
-| Audrey | 100 | 100 | 93.58333333333333 |
-| Vector Only | 41.66666666666667 | 25 | 0.25 |
+| Audrey | 100 | 100 | 14.75 |
+| Vector Only | 41.66666666666667 | 25 | 0.4166666666666667 |
 | Keyword + Recency | 41.66666666666667 | 25 | 0.5833333333333334 |
-| Recent Window | 37.5 | 25 | 0 |
+| Recent Window | 37.5 | 25 | 0.08333333333333333 |
 
 This output is a regression-gate result. The baselines are toy local baselines used to catch retrieval and lifecycle regressions in the Audrey codebase. They are not external systems, not tuned competitor implementations, and not GuardBench baselines (Ledger: E23-E24). The current suite covers retrieval and operation families such as information extraction, knowledge updates, multi-session reasoning, conflict resolution, procedural learning, privacy boundary, overwrite, delete-and-abstain, semantic merge, and procedural merge (Ledger: E23-E24).
 
@@ -55,7 +55,7 @@ It reports local adapters only, not external-system comparisons (Ledger: E46):
 | Evidence recall | 100% |
 | Redaction leaks | 0 |
 | Recall-degradation detection | 100% |
-| Guard latency p50 / p95 | 2.465 ms / 30.791 ms |
+| Guard latency p50 / p95 | 3.09 ms / 28.181 ms |
 | Published artifact raw-secret leaks | 0 |
 | Audrey Guard decision accuracy | 100% |
 | No-memory decision accuracy | 10% |