@ryuenn3123/agentic-senior-core 2.0.26 → 2.0.27

package/.agent-context/state/benchmark-history.json

@@ -0,0 +1,75 @@
+{
+  "generatedAt": "2026-04-17T03:20:15.400Z",
+  "reportName": "benchmark-history",
+  "maxEntries": 90,
+  "history": [
+    {
+      "generatedAt": "2026-04-17T02:54:01.239Z",
+      "releaseVersion": "2.0.26",
+      "fixtureCount": 12,
+      "top1Accuracy": 0.9167,
+      "manualCorrectionRate": 0.0833,
+      "benchmarkGatePassed": true,
+      "intelligencePassed": true,
+      "staleWatchlistCount": 0,
+      "reliabilityPassed": true,
+      "reliabilityRiskLevel": "monitor",
+      "incorrectDetectionRate": 0.0833,
+      "lowConfidenceRate": 0.0833,
+      "vulnerabilityTotal": null,
+      "criticalVulnerabilityCount": null,
+      "forbiddenContentPassed": true
+    },
+    {
+      "generatedAt": "2026-04-17T02:54:57.419Z",
+      "releaseVersion": "2.0.26",
+      "fixtureCount": 12,
+      "top1Accuracy": 0.9167,
+      "manualCorrectionRate": 0.0833,
+      "benchmarkGatePassed": true,
+      "intelligencePassed": true,
+      "staleWatchlistCount": 0,
+      "reliabilityPassed": true,
+      "reliabilityRiskLevel": "monitor",
+      "incorrectDetectionRate": 0.0833,
+      "lowConfidenceRate": 0.0833,
+      "vulnerabilityTotal": null,
+      "criticalVulnerabilityCount": null,
+      "forbiddenContentPassed": true
+    },
+    {
+      "generatedAt": "2026-04-17T03:19:31.047Z",
+      "releaseVersion": "2.0.26",
+      "fixtureCount": 12,
+      "top1Accuracy": 0.9167,
+      "manualCorrectionRate": 0.0833,
+      "benchmarkGatePassed": true,
+      "intelligencePassed": true,
+      "staleWatchlistCount": 0,
+      "reliabilityPassed": true,
+      "reliabilityRiskLevel": "monitor",
+      "incorrectDetectionRate": 0.0833,
+      "lowConfidenceRate": 0.0833,
+      "vulnerabilityTotal": null,
+      "criticalVulnerabilityCount": null,
+      "forbiddenContentPassed": true
+    },
+    {
+      "generatedAt": "2026-04-17T03:20:15.400Z",
+      "releaseVersion": "2.0.26",
+      "fixtureCount": 12,
+      "top1Accuracy": 0.9167,
+      "manualCorrectionRate": 0.0833,
+      "benchmarkGatePassed": true,
+      "intelligencePassed": true,
+      "staleWatchlistCount": 0,
+      "reliabilityPassed": true,
+      "reliabilityRiskLevel": "monitor",
+      "incorrectDetectionRate": 0.0833,
+      "lowConfidenceRate": 0.0833,
+      "vulnerabilityTotal": null,
+      "criticalVulnerabilityCount": null,
+      "forbiddenContentPassed": true
+    }
+  ]
+}

package/.agent-context/state/benchmark-trend-report.csv

@@ -0,0 +1,5 @@
+snapshotIndex,generatedAt,releaseVersion,top1Accuracy,manualCorrectionRate,incorrectDetectionRate,lowConfidenceRate,staleWatchlistCount,vulnerabilityTotal,criticalVulnerabilityCount,benchmarkGatePassed,intelligencePassed,reliabilityPassed,reliabilityRiskLevel
+"1","2026-04-17T02:54:01.239Z","2.0.26","0.9167","0.0833","0.0833","0.0833","0",,,"true","true","true","monitor"
+"2","2026-04-17T02:54:57.419Z","2.0.26","0.9167","0.0833","0.0833","0.0833","0",,,"true","true","true","monitor"
+"3","2026-04-17T03:19:31.047Z","2.0.26","0.9167","0.0833","0.0833","0.0833","0",,,"true","true","true","monitor"
+"4","2026-04-17T03:20:15.400Z","2.0.26","0.9167","0.0833","0.0833","0.0833","0",,,"true","true","true","monitor"

package/.agent-context/state/benchmark-trend-report.json

@@ -0,0 +1,140 @@
+{
+  "generatedAt": "2026-04-17T03:20:15.400Z",
+  "reportName": "benchmark-trend-report",
+  "releaseVersion": "2.0.26",
+  "historyCount": 4,
+  "releaseDelta": {
+    "currentReleaseVersion": "2.0.26",
+    "previousReleaseVersion": "2.0.26",
+    "comparedSnapshot": {
+      "currentGeneratedAt": "2026-04-17T03:20:15.400Z",
+      "previousGeneratedAt": "2026-04-17T03:19:31.047Z"
+    },
+    "top1AccuracyDelta": 0,
+    "manualCorrectionRateDelta": 0,
+    "staleWatchlistCountDelta": 0,
+    "vulnerabilityTotalDelta": 0,
+    "summary": [
+      "top1Accuracy: +0",
+      "manualCorrectionRate: +0",
+      "staleWatchlistCount: +0",
+      "vulnerabilityTotal: +0"
+    ]
+  },
+  "trendTable": [
+    {
+      "snapshotIndex": 1,
+      "generatedAt": "2026-04-17T02:54:01.239Z",
+      "releaseVersion": "2.0.26",
+      "top1Accuracy": 0.9167,
+      "manualCorrectionRate": 0.0833,
+      "incorrectDetectionRate": 0.0833,
+      "lowConfidenceRate": 0.0833,
+      "staleWatchlistCount": 0,
+      "vulnerabilityTotal": null,
+      "criticalVulnerabilityCount": null,
+      "benchmarkGatePassed": true,
+      "intelligencePassed": true,
+      "reliabilityPassed": true,
+      "reliabilityRiskLevel": "monitor"
+    },
+    {
+      "snapshotIndex": 2,
+      "generatedAt": "2026-04-17T02:54:57.419Z",
+      "releaseVersion": "2.0.26",
+      "top1Accuracy": 0.9167,
+      "manualCorrectionRate": 0.0833,
+      "incorrectDetectionRate": 0.0833,
+      "lowConfidenceRate": 0.0833,
+      "staleWatchlistCount": 0,
+      "vulnerabilityTotal": null,
+      "criticalVulnerabilityCount": null,
+      "benchmarkGatePassed": true,
+      "intelligencePassed": true,
+      "reliabilityPassed": true,
+      "reliabilityRiskLevel": "monitor"
+    },
+    {
+      "snapshotIndex": 3,
+      "generatedAt": "2026-04-17T03:19:31.047Z",
+      "releaseVersion": "2.0.26",
+      "top1Accuracy": 0.9167,
+      "manualCorrectionRate": 0.0833,
+      "incorrectDetectionRate": 0.0833,
+      "lowConfidenceRate": 0.0833,
+      "staleWatchlistCount": 0,
+      "vulnerabilityTotal": null,
+      "criticalVulnerabilityCount": null,
+      "benchmarkGatePassed": true,
+      "intelligencePassed": true,
+      "reliabilityPassed": true,
+      "reliabilityRiskLevel": "monitor"
+    },
+    {
+      "snapshotIndex": 4,
+      "generatedAt": "2026-04-17T03:20:15.400Z",
+      "releaseVersion": "2.0.26",
+      "top1Accuracy": 0.9167,
+      "manualCorrectionRate": 0.0833,
+      "incorrectDetectionRate": 0.0833,
+      "lowConfidenceRate": 0.0833,
+      "staleWatchlistCount": 0,
+      "vulnerabilityTotal": null,
+      "criticalVulnerabilityCount": null,
+      "benchmarkGatePassed": true,
+      "intelligencePassed": true,
+      "reliabilityPassed": true,
+      "reliabilityRiskLevel": "monitor"
+    }
+  ],
+  "chartSeries": {
+    "generatedAt": [
+      "2026-04-17T02:54:01.239Z",
+      "2026-04-17T02:54:57.419Z",
+      "2026-04-17T03:19:31.047Z",
+      "2026-04-17T03:20:15.400Z"
+    ],
+    "top1Accuracy": [
+      0.9167,
+      0.9167,
+      0.9167,
+      0.9167
+    ],
+    "manualCorrectionRate": [
+      0.0833,
+      0.0833,
+      0.0833,
+      0.0833
+    ],
+    "incorrectDetectionRate": [
+      0.0833,
+      0.0833,
+      0.0833,
+      0.0833
+    ],
+    "lowConfidenceRate": [
+      0.0833,
+      0.0833,
+      0.0833,
+      0.0833
+    ],
+    "staleWatchlistCount": [
+      0,
+      0,
+      0,
+      0
+    ],
+    "vulnerabilityTotal": [
+      null,
+      null,
+      null,
+      null
+    ]
+  },
+  "artifacts": {
+    "historyPath": ".agent-context/state/benchmark-history.json",
+    "jsonPath": ".agent-context/state/benchmark-trend-report.json",
+    "csvPath": ".agent-context/state/benchmark-trend-report.csv",
+    "writeMode": "stdout-and-file"
+  }
+}

package/.agent-context/state/benchmark-watchlist.json

@@ -3,17 +3,17 @@
     {
       "repository": "sickn33/antigravity-awesome-skills",
       "owner": "core-architecture",
-      "lastReviewedAt": "2026-04-02"
+      "lastReviewedAt": "2026-04-17"
     },
     {
       "repository": "github/awesome-copilot",
       "owner": "core-architecture",
-      "lastReviewedAt": "2026-04-02"
+      "lastReviewedAt": "2026-04-17"
     },
     {
       "repository": "MiniMax-AI/skills",
       "owner": "frontend-governance",
-      "lastReviewedAt": "2026-04-02"
+      "lastReviewedAt": "2026-04-17"
     }
   ]
 }

package/.agent-context/state/memory-adapter-contract.json

@@ -0,0 +1,52 @@
+{
+  "schemaVersion": "1.0.0",
+  "contractName": "cross-agent-memory-adapter",
+  "description": "Adapter contract for ingesting and retrieving shared memory observations across IDE hosts.",
+  "requiredAdapters": [
+    "claude-code",
+    "gemini-cli",
+    "vscode-chat"
+  ],
+  "requiredOperations": {
+    "ingestion": [
+      "captureObservation",
+      "captureSessionSummary"
+    ],
+    "retrieval": [
+      "searchIndex",
+      "getTimeline",
+      "getObservations"
+    ],
+    "privacy": [
+      "applyPrivateTagRedaction",
+      "applyInlineSecretRedaction"
+    ]
+  },
+  "adapters": [
+    {
+      "adapterId": "claude-code",
+      "hostType": "plugin-hooks",
+      "status": "pilot-ready",
+      "ingestionEvents": ["SessionStart", "UserPromptSubmit", "PostToolUse", "Stop", "SessionEnd"],
+      "retrievalMode": "mcp-tools"
+    },
+    {
+      "adapterId": "gemini-cli",
+      "hostType": "plugin-hooks",
+      "status": "pilot-ready",
+      "ingestionEvents": ["session_start", "prompt_submit", "post_tool", "session_end"],
+      "retrievalMode": "mcp-tools"
+    },
+    {
+      "adapterId": "vscode-chat",
+      "hostType": "chat-customization-plugin",
+      "status": "pilot-ready",
+      "ingestionEvents": ["chatStart", "promptSubmit", "postToolUse", "chatEnd"],
+      "retrievalMode": "mcp-tools"
+    }
+  ],
+  "notes": [
+    "Web chat hosts are explicitly out of scope for this pilot because local runtime hooks are unavailable.",
+    "Adapters should emit provider-agnostic payloads matching .agent-context/state/memory-schema-v1.json."
+  ]
+}

package/.agent-context/state/memory-continuity-benchmark.json

@@ -0,0 +1,132 @@
+{
+  "generatedAt": "2026-04-17T03:27:46.024Z",
+  "reportName": "memory-continuity-benchmark",
+  "schemaVersion": "1.0.0",
+  "passed": true,
+  "failureCount": 0,
+  "thresholds": {
+    "minimumRelevantRecall": 0.8,
+    "minimumSessionStartTokenSavingsPercent": 35,
+    "maximumUnsafeObservationCount": 0
+  },
+  "adapterCoverage": {
+    "requiredAdapterIds": [
+      "claude-code",
+      "gemini-cli",
+      "vscode-chat"
+    ],
+    "availableAdapterIds": [
+      "claude-code",
+      "gemini-cli",
+      "vscode-chat"
+    ],
+    "missingAdapterIds": [],
+    "passed": true
+  },
+  "privacyControls": {
+    "redactedObservationCount": 2,
+    "privateTagRedactionCount": 1,
+    "inlineRedactionCount": 1,
+    "unsafeObservationCount": 0
+  },
+  "continuitySummary": {
+    "totalObservationCount": 5,
+    "scenarioCount": 3,
+    "averageRelevantRecall": 1,
+    "averageSessionStartTokenSavingsPercent": 63.17
+  },
+  "scenarios": [
+    {
+      "scenarioId": "docker-lane-hydration",
+      "query": "what is docker strategy for development and production",
+      "expectedObservationIds": [
+        "obs-001"
+      ],
+      "indexObservationIds": [
+        "obs-001",
+        "obs-005",
+        "obs-003",
+        "obs-002",
+        "obs-004"
+      ],
+      "hydratedObservationIds": [
+        "obs-001"
+      ],
+      "relevantRecall": 1,
+      "fullContextTokenEstimate": 267,
+      "sessionStartTokenEstimate": 103,
+      "sessionStartTokenSavingsPercent": 61.42
+    },
+    {
+      "scenarioId": "runtime-hydration",
+      "query": "which runtime target should we prefer on windows with wsl",
+      "expectedObservationIds": [
+        "obs-002"
+      ],
+      "indexObservationIds": [
+        "obs-002",
+        "obs-001",
+        "obs-005",
+        "obs-004",
+        "obs-003"
+      ],
+      "hydratedObservationIds": [
+        "obs-002"
+      ],
+      "relevantRecall": 1,
+      "fullContextTokenEstimate": 267,
+      "sessionStartTokenEstimate": 97,
+      "sessionStartTokenSavingsPercent": 63.67
+    },
+    {
+      "scenarioId": "frontend-quality-hydration",
+      "query": "show frontend rubric quality decisions",
+      "expectedObservationIds": [
+        "obs-003"
+      ],
+      "indexObservationIds": [
+        "obs-003",
+        "obs-005",
+        "obs-004",
+        "obs-002",
+        "obs-001"
+      ],
+      "hydratedObservationIds": [
+        "obs-003"
+      ],
+      "relevantRecall": 1,
+      "fullContextTokenEstimate": 267,
+      "sessionStartTokenEstimate": 95,
+      "sessionStartTokenSavingsPercent": 64.42
+    }
+  ],
+  "references": {
+    "memorySchemaPath": ".agent-context/state/memory-schema-v1.json",
+    "memoryAdapterContractPath": ".agent-context/state/memory-adapter-contract.json",
+    "benchmarkOutputPath": ".agent-context/state/memory-continuity-benchmark.json",
+    "schemaDeclaredVersion": "1.0.0",
+    "adapterContractVersion": "1.0.0"
+  },
+  "checks": [
+    {
+      "checkName": "adapter-coverage",
+      "passed": true,
+      "details": "required=3 missing=0"
+    },
+    {
+      "checkName": "continuity-recall-threshold",
+      "passed": true,
+      "details": "averageRelevantRecall=1 minimum=0.8"
+    },
+    {
+      "checkName": "session-start-token-savings-threshold",
+      "passed": true,
+      "details": "averageSessionStartTokenSavingsPercent=63.17 minimum=35"
+    },
+    {
+      "checkName": "privacy-redaction-safety",
+      "passed": true,
+      "details": "unsafeObservationCount=0 max=0"
+    }
+  ]
+}

package/.agent-context/state/memory-schema-v1.json

@@ -0,0 +1,88 @@
+{
+  "schemaVersion": "1.0.0",
+  "schemaName": "cross-agent-memory-observation",
+  "description": "Provider-agnostic schema for persistent memory observations shared across coding agents and IDE hosts.",
+  "requiredFields": [
+    "id",
+    "projectId",
+    "sessionId",
+    "adapterId",
+    "eventType",
+    "timestamp",
+    "title",
+    "summary",
+    "detail",
+    "privacy"
+  ],
+  "fieldDefinitions": {
+    "id": {
+      "type": "string",
+      "description": "Stable unique observation identifier."
+    },
+    "projectId": {
+      "type": "string",
+      "description": "Repository or workspace identifier."
+    },
+    "sessionId": {
+      "type": "string",
+      "description": "Source session identifier from host adapter."
+    },
+    "adapterId": {
+      "type": "string",
+      "allowedValues": ["claude-code", "gemini-cli", "vscode-chat", "custom"],
+      "description": "Host adapter that captured this observation."
+    },
+    "eventType": {
+      "type": "string",
+      "allowedValues": ["prompt", "tool-use", "decision", "summary", "issue", "context"],
+      "description": "Observation type for retrieval filtering."
+    },
+    "timestamp": {
+      "type": "string",
+      "format": "date-time",
+      "description": "ISO timestamp when observation was captured."
+    },
+    "title": {
+      "type": "string",
+      "description": "Compact human-readable headline."
+    },
+    "summary": {
+      "type": "string",
+      "description": "Compact session-start payload used for progressive disclosure."
+    },
+    "detail": {
+      "type": "string",
+      "description": "Expanded observation text fetched on demand."
+    },
+    "tags": {
+      "type": "array",
+      "items": "string",
+      "description": "Optional normalized tags for query refinement."
+    },
+    "privacy": {
+      "type": "object",
+      "requiredFields": ["level", "redactionApplied"],
+      "fieldDefinitions": {
+        "level": {
+          "type": "string",
+          "allowedValues": ["public", "internal", "restricted"],
+          "description": "Privacy classification level."
+        },
+        "redactionApplied": {
+          "type": "boolean",
+          "description": "Indicates whether privacy sanitization modified payload content."
+        },
+        "redactionReasons": {
+          "type": "array",
+          "items": "string",
+          "description": "Redaction reason tags such as private-tag or token-like-value."
+        }
+      }
+    }
+  },
+  "retrievalContract": {
+    "sessionStartPayload": ["id", "adapterId", "eventType", "timestamp", "title", "summary"],
+    "onDemandPayload": ["detail", "tags", "privacy"],
+    "progressiveDisclosure": true
+  }
+}

package/.cursorrules

@@ -1,6 +1,6 @@
 # AGENTIC-SENIOR-CORE DYNAMIC GOVERNANCE RULESET
 
-Generated by Agentic-Senior-Core CLI v2.0.26
+Generated by Agentic-Senior-Core CLI v2.0.27
 Timestamp: 2026-04-15T00:14:51.184Z
 Selected profile: beginner
 Selected policy file: .agent-context/policies/llm-judge-threshold.json

package/.windsurfrules

@@ -1,6 +1,6 @@
 # AGENTIC-SENIOR-CORE DYNAMIC GOVERNANCE RULESET
 
-Generated by Agentic-Senior-Core CLI v2.0.26
+Generated by Agentic-Senior-Core CLI v2.0.27
 Timestamp: 2026-04-15T00:14:51.184Z
 Selected profile: beginner
 Selected policy file: .agent-context/policies/llm-judge-threshold.json

package/README.md

@@ -294,6 +294,27 @@ For CI pipelines that only need stdout JSON:
 node ./scripts/benchmark-writer-judge-matrix.mjs --stdout-only
 ```
 
+### Memory Continuity Benchmark (V2.5.3 Pilot)
+
+Validate cross-agent memory hydration quality, privacy redaction safety, and session-start token savings:
+
+```bash
+npm run benchmark:continuity
+```
+
+This command writes:
+- `.agent-context/state/memory-continuity-benchmark.json`
+
+Reference artifacts:
+- `.agent-context/state/memory-schema-v1.json`
+- `.agent-context/state/memory-adapter-contract.json`
+
+For CI pipelines that only need stdout JSON:
+
+```bash
+node ./scripts/memory-continuity-benchmark.mjs --stdout-only
+```
+
 ### Benchmark Quickstart Path (V2.5)
 
 For new users, run this minimal sequence first:
@@ -302,6 +323,7 @@ For new users, run this minimal sequence first:
 npm run benchmark:detection
 npm run benchmark:writer-judge
 npm run benchmark:bundle
+npm run benchmark:continuity
 ```
 
 This gives a fast baseline of accuracy, writer-judge comparison, and evidence packaging in one pass.