memorix 1.7.1 → 1.7.2

package/CHANGELOG.md

@@ -4,6 +4,22 @@ All notable changes to this project will be documented in this file.
 
 ## [Unreleased]
 
+## [1.7.2] - 2026-08-19
+
+### Changed
+- **Faster large-store cold retrieval** -- startup hydration now inserts
+  persisted observations into Orama in bounded batches. This avoids rebuilding
+  index balance once per row while continuing to yield to HTTP health traffic.
+- **Fail-closed large-store evidence** -- the reproducible 40,000-record gate
+  now exercises and times HTTP readiness, MCP initialize/bind/context/search/
+  store, hook capture, and SDK reopen behavior with hard request and release
+  budgets instead of checking persistence alone.
+- **Isolated index tests** -- startup hydration tests explicitly disable the
+  embedding lane and restore the caller environment instead of reading an
+  operator's provider configuration or local vector cache.
+- **Current public guidance** -- README and canonical development/status docs
+  now describe preview-first Dashboard maintenance and the 1.7 product line.
+
 ## [1.7.1] - 2026-08-19
 
 ### Fixed

package/README.md

@@ -80,7 +80,7 @@ Memorix is more than a memory store. It also installs agent integrations, keeps
 | Agent setup | One setup path for MCP, rules, hooks, skills, plugins, bundles, or extensions depending on the agent | `memorix setup --agent <agent>` |
 | Agent doctor | Checks whether agent MCP config and guidance are current, then repairs Memorix-owned entries when needed | `memorix doctor agents`, `memorix repair agents` |
 | Hooks and skills | Optional capture from supported agents, plus reusable project skills promoted from durable knowledge | `memorix hooks`, `memorix skills` |
-| Dashboard and HTTP | A local web UI and shared MCP endpoint for browsing memory, project state, teams, and diagnostics | `memorix dashboard`, `memorix background start` |
+| Dashboard and HTTP | A local web UI and shared MCP endpoint for browsing memory, project state, teams, diagnostics, and preview-first cleanup, consolidation, deduplication, and retention actions | `memorix dashboard`, `memorix background start` |
 | Orchestration and team work | Task planning, worker handoffs, file locks, messages, verification gates, and review loops | `memorix orchestrate`, `memorix team`, `memorix lock` |
 | memcode | A bundled terminal coding agent that already reads and writes the same project memory | `memorix`, `memcode` |
 | CLI and SDK | Scriptable access for automation, imports/exports, diagnostics, and custom integrations | `memorix ...`, `createMemoryClient()` |

package/README.zh-CN.md

@@ -79,7 +79,7 @@ Memorix 不只是一个记忆库。它还负责安装 Agent 接入、保留有
 | Agent setup | 按目标 Agent 写入 MCP、rules、hooks、skills、plugin、bundle 或 extension | `memorix setup --agent <agent>` |
 | Agent doctor | 检查 Agent 的 MCP 配置和规则是否是当前版本，并修复 Memorix 自己管理的条目 | `memorix doctor agents`、`memorix repair agents` |
 | Hooks 和 skills | 在支持的 Agent 中可选捕获工作事件，并把稳定知识提升成可复用项目技能 | `memorix hooks`、`memorix skills` |
-| Dashboard 和 HTTP | 本地 Web UI 与共享 MCP endpoint，用于浏览记忆、项目状态、团队和诊断信息 | `memorix dashboard`、`memorix background start` |
+| Dashboard 和 HTTP | 本地 Web UI 与共享 MCP endpoint，用于浏览记忆、项目状态、团队和诊断，并在预览确认后执行清理、合并、去重和保留归档 | `memorix dashboard`、`memorix background start` |
 | Orchestration 和团队协作 | 任务规划、worker 交接、文件锁、消息、验证门和 review loop | `memorix orchestrate`、`memorix team`、`memorix lock` |
 | memcode | 内置终端 Coding Agent，默认读写同一套项目记忆 | `memorix`、`memcode` |
 | CLI 和 SDK | 给自动化、导入导出、诊断和自定义集成使用的本地接口 | `memorix ...`、`createMemoryClient()` |

package/dist/cli/index.js

@@ -3232,7 +3232,7 @@ ${import_picocolors.default.gray(m3)}  ${s4}
 
 // src/cli/version.ts
 function getCliVersion() {
-  return true ? "1.7.1" : pkg.version;
+  return true ? "1.7.2" : pkg.version;
 }
 var init_version = __esm({
   "src/cli/version.ts"() {
@@ -185335,7 +185335,7 @@ async function hydrateIndex(observations2, options = {}) {
   }
   const deferCachedVectors = options.deferCachedVectors === true && Boolean(indexEmbeddingProvider?.getCachedEmbeddings);
   const cachedVectors = deferCachedVectors ? candidates.map(() => null) : await getCachedEmbeddings(candidates.map(({ observation }) => observationEmbeddingText(observation)));
-  let inserted = 0;
+  const documents = [];
   for (let index = 0; index < candidates.length; index++) {
     const { observation: obs, id } = candidates[index];
     try {
@@ -185369,17 +185369,34 @@ async function hydrateIndex(observations2, options = {}) {
         knowledgeLayer: resolveKnowledgeLayer("observation", obs.sourceDetail, obs.source),
         ...compatibleVector ? { embedding: compatibleVector } : {}
       };
-      await insert3(database, doc);
-      rememberObservationDoc(doc);
-      inserted++;
-      if (inserted % 50 === 0) {
-        await new Promise((resolve6) => setImmediate(resolve6));
-      }
-      if (inserted % 200 === 0) {
-        await new Promise((resolve6) => setTimeout(resolve6, 0));
-      }
+      documents.push(doc);
+    } catch {
+    }
+  }
+  let inserted = 0;
+  const hydrationBatchSize = 200;
+  for (let start = 0; start < documents.length; start += hydrationBatchSize) {
+    const batch = documents.slice(start, start + hydrationBatchSize);
+    try {
+      await insertMultiple(database, batch, batch.length);
+      for (const doc of batch) rememberObservationDoc(doc);
+      inserted += batch.length;
     } catch {
+      for (const doc of batch) {
+        if (getByID(database, doc.id)) {
+          rememberObservationDoc(doc);
+          inserted++;
+          continue;
+        }
+        try {
+          await insert3(database, doc);
+          rememberObservationDoc(doc);
+          inserted++;
+        } catch {
+        }
+      }
     }
+    await new Promise((resolve6) => setImmediate(resolve6));
   }
   deferredCachedVectorHydration = options.skipCachedVectors ? null : deferCachedVectors ? attachCachedVectors(database, candidates).catch(() => {
   }) : null;
@@ -260203,7 +260220,7 @@ The path should point to a directory containing a .git folder.`
   };
   const server = existingServer ?? new McpServer({
     name: "memorix",
-    version: true ? "1.7.1" : "1.0.1"
+    version: true ? "1.7.2" : "1.0.1"
   });
   const originalRegisterTool = server.registerTool.bind(server);
   server.registerTool = ((name, ...args) => {

package/dist/index.js

@@ -6888,7 +6888,7 @@ __export(orama_store_exports, {
   searchObservations: () => searchObservations,
   updateObservationMetadata: () => updateObservationMetadata
 });
-import { create, insert as insert2, search, remove as remove2, update, count, getByID } from "@orama/orama";
+import { create, insert as insert2, insertMultiple, search, remove as remove2, update, count, getByID } from "@orama/orama";
 function getLastSearchMode(projectId) {
   return lastSearchModeByProject.get(projectId ?? SEARCH_MODE_DEFAULT_KEY) ?? "fulltext";
 }
@@ -7100,7 +7100,7 @@ async function hydrateIndex(observations2, options = {}) {
   }
   const deferCachedVectors = options.deferCachedVectors === true && Boolean(indexEmbeddingProvider?.getCachedEmbeddings);
   const cachedVectors = deferCachedVectors ? candidates.map(() => null) : await getCachedEmbeddings(candidates.map(({ observation }) => observationEmbeddingText(observation)));
-  let inserted = 0;
+  const documents = [];
   for (let index = 0; index < candidates.length; index++) {
     const { observation: obs, id } = candidates[index];
     try {
@@ -7134,17 +7134,34 @@ async function hydrateIndex(observations2, options = {}) {
         knowledgeLayer: resolveKnowledgeLayer("observation", obs.sourceDetail, obs.source),
         ...compatibleVector ? { embedding: compatibleVector } : {}
       };
-      await insert2(database, doc);
-      rememberObservationDoc(doc);
-      inserted++;
-      if (inserted % 50 === 0) {
-        await new Promise((resolve2) => setImmediate(resolve2));
-      }
-      if (inserted % 200 === 0) {
-        await new Promise((resolve2) => setTimeout(resolve2, 0));
-      }
+      documents.push(doc);
+    } catch {
+    }
+  }
+  let inserted = 0;
+  const hydrationBatchSize = 200;
+  for (let start = 0; start < documents.length; start += hydrationBatchSize) {
+    const batch = documents.slice(start, start + hydrationBatchSize);
+    try {
+      await insertMultiple(database, batch, batch.length);
+      for (const doc of batch) rememberObservationDoc(doc);
+      inserted += batch.length;
     } catch {
+      for (const doc of batch) {
+        if (getByID(database, doc.id)) {
+          rememberObservationDoc(doc);
+          inserted++;
+          continue;
+        }
+        try {
+          await insert2(database, doc);
+          rememberObservationDoc(doc);
+          inserted++;
+        } catch {
+        }
+      }
     }
+    await new Promise((resolve2) => setImmediate(resolve2));
   }
   deferredCachedVectorHydration = options.skipCachedVectors ? null : deferCachedVectors ? attachCachedVectors(database, candidates).catch(() => {
   }) : null;
@@ -167285,7 +167302,7 @@ The path should point to a directory containing a .git folder.`
   };
   const server = existingServer ?? new McpServer({
     name: "memorix",
-    version: true ? "1.7.1" : "1.0.1"
+    version: true ? "1.7.2" : "1.0.1"
   });
   const originalRegisterTool = server.registerTool.bind(server);
   server.registerTool = ((name, ...args) => {
@@ -171725,7 +171742,7 @@ var init_serve_shared = __esm({
 
 // src/cli/version.ts
 function getCliVersion() {
-  return true ? "1.7.1" : pkg.version;
+  return true ? "1.7.2" : pkg.version;
 }
 var init_version3 = __esm({
   "src/cli/version.ts"() {

package/dist/maintenance-runner.js

@@ -6094,7 +6094,7 @@ __export(orama_store_exports, {
   searchObservations: () => searchObservations,
   updateObservationMetadata: () => updateObservationMetadata
 });
-import { create, insert as insert2, search, remove as remove2, update, count, getByID } from "@orama/orama";
+import { create, insert as insert2, insertMultiple, search, remove as remove2, update, count, getByID } from "@orama/orama";
 function getLastSearchMode(projectId) {
   return lastSearchModeByProject.get(projectId ?? SEARCH_MODE_DEFAULT_KEY) ?? "fulltext";
 }
@@ -6306,7 +6306,7 @@ async function hydrateIndex(observations2, options = {}) {
   }
   const deferCachedVectors = options.deferCachedVectors === true && Boolean(indexEmbeddingProvider?.getCachedEmbeddings);
   const cachedVectors = deferCachedVectors ? candidates.map(() => null) : await getCachedEmbeddings(candidates.map(({ observation }) => observationEmbeddingText(observation)));
-  let inserted = 0;
+  const documents = [];
   for (let index = 0; index < candidates.length; index++) {
     const { observation: obs, id } = candidates[index];
     try {
@@ -6340,17 +6340,34 @@ async function hydrateIndex(observations2, options = {}) {
         knowledgeLayer: resolveKnowledgeLayer("observation", obs.sourceDetail, obs.source),
         ...compatibleVector ? { embedding: compatibleVector } : {}
       };
-      await insert2(database, doc);
-      rememberObservationDoc(doc);
-      inserted++;
-      if (inserted % 50 === 0) {
-        await new Promise((resolve) => setImmediate(resolve));
-      }
-      if (inserted % 200 === 0) {
-        await new Promise((resolve) => setTimeout(resolve, 0));
-      }
+      documents.push(doc);
+    } catch {
+    }
+  }
+  let inserted = 0;
+  const hydrationBatchSize = 200;
+  for (let start = 0; start < documents.length; start += hydrationBatchSize) {
+    const batch = documents.slice(start, start + hydrationBatchSize);
+    try {
+      await insertMultiple(database, batch, batch.length);
+      for (const doc of batch) rememberObservationDoc(doc);
+      inserted += batch.length;
     } catch {
+      for (const doc of batch) {
+        if (getByID(database, doc.id)) {
+          rememberObservationDoc(doc);
+          inserted++;
+          continue;
+        }
+        try {
+          await insert2(database, doc);
+          rememberObservationDoc(doc);
+          inserted++;
+        } catch {
+        }
+      }
     }
+    await new Promise((resolve) => setImmediate(resolve));
   }
   deferredCachedVectorHydration = options.skipCachedVectors ? null : deferCachedVectors ? attachCachedVectors(database, candidates).catch(() => {
   }) : null;

package/dist/media-audio-runner.js

@@ -6580,7 +6580,7 @@ __export(orama_store_exports, {
   searchObservations: () => searchObservations,
   updateObservationMetadata: () => updateObservationMetadata
 });
-import { create, insert as insert2, search, remove as remove2, update, count, getByID } from "@orama/orama";
+import { create, insert as insert2, insertMultiple, search, remove as remove2, update, count, getByID } from "@orama/orama";
 function getLastSearchMode(projectId) {
   return lastSearchModeByProject.get(projectId ?? SEARCH_MODE_DEFAULT_KEY) ?? "fulltext";
 }
@@ -6792,7 +6792,7 @@ async function hydrateIndex(observations2, options = {}) {
   }
   const deferCachedVectors = options.deferCachedVectors === true && Boolean(indexEmbeddingProvider?.getCachedEmbeddings);
   const cachedVectors = deferCachedVectors ? candidates.map(() => null) : await getCachedEmbeddings(candidates.map(({ observation }) => observationEmbeddingText(observation)));
-  let inserted = 0;
+  const documents = [];
   for (let index = 0; index < candidates.length; index++) {
     const { observation: obs, id } = candidates[index];
     try {
@@ -6826,17 +6826,34 @@ async function hydrateIndex(observations2, options = {}) {
         knowledgeLayer: resolveKnowledgeLayer("observation", obs.sourceDetail, obs.source),
         ...compatibleVector ? { embedding: compatibleVector } : {}
       };
-      await insert2(database, doc);
-      rememberObservationDoc(doc);
-      inserted++;
-      if (inserted % 50 === 0) {
-        await new Promise((resolve) => setImmediate(resolve));
-      }
-      if (inserted % 200 === 0) {
-        await new Promise((resolve) => setTimeout(resolve, 0));
-      }
+      documents.push(doc);
+    } catch {
+    }
+  }
+  let inserted = 0;
+  const hydrationBatchSize = 200;
+  for (let start = 0; start < documents.length; start += hydrationBatchSize) {
+    const batch = documents.slice(start, start + hydrationBatchSize);
+    try {
+      await insertMultiple(database, batch, batch.length);
+      for (const doc of batch) rememberObservationDoc(doc);
+      inserted += batch.length;
     } catch {
+      for (const doc of batch) {
+        if (getByID(database, doc.id)) {
+          rememberObservationDoc(doc);
+          inserted++;
+          continue;
+        }
+        try {
+          await insert2(database, doc);
+          rememberObservationDoc(doc);
+          inserted++;
+        } catch {
+        }
+      }
     }
+    await new Promise((resolve) => setImmediate(resolve));
   }
   deferredCachedVectorHydration = options.skipCachedVectors ? null : deferCachedVectors ? attachCachedVectors(database, candidates).catch(() => {
   }) : null;

package/dist/media-video-runner.js

@@ -5597,7 +5597,7 @@ __export(orama_store_exports, {
   searchObservations: () => searchObservations,
   updateObservationMetadata: () => updateObservationMetadata
 });
-import { create, insert as insert2, search, remove as remove2, update, count, getByID } from "@orama/orama";
+import { create, insert as insert2, insertMultiple, search, remove as remove2, update, count, getByID } from "@orama/orama";
 function getLastSearchMode(projectId) {
   return lastSearchModeByProject.get(projectId ?? SEARCH_MODE_DEFAULT_KEY) ?? "fulltext";
 }
@@ -5809,7 +5809,7 @@ async function hydrateIndex(observations2, options = {}) {
   }
   const deferCachedVectors = options.deferCachedVectors === true && Boolean(indexEmbeddingProvider?.getCachedEmbeddings);
   const cachedVectors = deferCachedVectors ? candidates.map(() => null) : await getCachedEmbeddings(candidates.map(({ observation }) => observationEmbeddingText(observation)));
-  let inserted = 0;
+  const documents = [];
   for (let index = 0; index < candidates.length; index++) {
     const { observation: obs, id } = candidates[index];
     try {
@@ -5843,17 +5843,34 @@ async function hydrateIndex(observations2, options = {}) {
         knowledgeLayer: resolveKnowledgeLayer("observation", obs.sourceDetail, obs.source),
         ...compatibleVector ? { embedding: compatibleVector } : {}
       };
-      await insert2(database, doc);
-      rememberObservationDoc(doc);
-      inserted++;
-      if (inserted % 50 === 0) {
-        await new Promise((resolve) => setImmediate(resolve));
-      }
-      if (inserted % 200 === 0) {
-        await new Promise((resolve) => setTimeout(resolve, 0));
-      }
+      documents.push(doc);
+    } catch {
+    }
+  }
+  let inserted = 0;
+  const hydrationBatchSize = 200;
+  for (let start = 0; start < documents.length; start += hydrationBatchSize) {
+    const batch = documents.slice(start, start + hydrationBatchSize);
+    try {
+      await insertMultiple(database, batch, batch.length);
+      for (const doc of batch) rememberObservationDoc(doc);
+      inserted += batch.length;
     } catch {
+      for (const doc of batch) {
+        if (getByID(database, doc.id)) {
+          rememberObservationDoc(doc);
+          inserted++;
+          continue;
+        }
+        try {
+          await insert2(database, doc);
+          rememberObservationDoc(doc);
+          inserted++;
+        } catch {
+        }
+      }
     }
+    await new Promise((resolve) => setImmediate(resolve));
   }
   deferredCachedVectorHydration = options.skipCachedVectors ? null : deferCachedVectors ? attachCachedVectors(database, candidates).catch(() => {
   }) : null;

package/dist/memcode-runtime/CHANGELOG.md

@@ -4,6 +4,22 @@ All notable changes to this project will be documented in this file.
 
 ## [Unreleased]
 
+## [1.7.2] - 2026-08-19
+
+### Changed
+- **Faster large-store cold retrieval** -- startup hydration now inserts
+  persisted observations into Orama in bounded batches. This avoids rebuilding
+  index balance once per row while continuing to yield to HTTP health traffic.
+- **Fail-closed large-store evidence** -- the reproducible 40,000-record gate
+  now exercises and times HTTP readiness, MCP initialize/bind/context/search/
+  store, hook capture, and SDK reopen behavior with hard request and release
+  budgets instead of checking persistence alone.
+- **Isolated index tests** -- startup hydration tests explicitly disable the
+  embedding lane and restore the caller environment instead of reading an
+  operator's provider configuration or local vector cache.
+- **Current public guidance** -- README and canonical development/status docs
+  now describe preview-first Dashboard maintenance and the 1.7 product line.
+
 ## [1.7.1] - 2026-08-19
 
 ### Fixed

package/dist/sdk.js

@@ -6876,7 +6876,7 @@ __export(orama_store_exports, {
   searchObservations: () => searchObservations,
   updateObservationMetadata: () => updateObservationMetadata
 });
-import { create, insert as insert2, search, remove as remove2, update, count, getByID } from "@orama/orama";
+import { create, insert as insert2, insertMultiple, search, remove as remove2, update, count, getByID } from "@orama/orama";
 function getLastSearchMode(projectId) {
   return lastSearchModeByProject.get(projectId ?? SEARCH_MODE_DEFAULT_KEY) ?? "fulltext";
 }
@@ -7088,7 +7088,7 @@ async function hydrateIndex(observations2, options = {}) {
   }
   const deferCachedVectors = options.deferCachedVectors === true && Boolean(indexEmbeddingProvider?.getCachedEmbeddings);
   const cachedVectors = deferCachedVectors ? candidates.map(() => null) : await getCachedEmbeddings(candidates.map(({ observation }) => observationEmbeddingText(observation)));
-  let inserted = 0;
+  const documents = [];
   for (let index = 0; index < candidates.length; index++) {
     const { observation: obs, id } = candidates[index];
     try {
@@ -7122,17 +7122,34 @@ async function hydrateIndex(observations2, options = {}) {
         knowledgeLayer: resolveKnowledgeLayer("observation", obs.sourceDetail, obs.source),
         ...compatibleVector ? { embedding: compatibleVector } : {}
       };
-      await insert2(database, doc);
-      rememberObservationDoc(doc);
-      inserted++;
-      if (inserted % 50 === 0) {
-        await new Promise((resolve2) => setImmediate(resolve2));
-      }
-      if (inserted % 200 === 0) {
-        await new Promise((resolve2) => setTimeout(resolve2, 0));
-      }
+      documents.push(doc);
+    } catch {
+    }
+  }
+  let inserted = 0;
+  const hydrationBatchSize = 200;
+  for (let start = 0; start < documents.length; start += hydrationBatchSize) {
+    const batch = documents.slice(start, start + hydrationBatchSize);
+    try {
+      await insertMultiple(database, batch, batch.length);
+      for (const doc of batch) rememberObservationDoc(doc);
+      inserted += batch.length;
     } catch {
+      for (const doc of batch) {
+        if (getByID(database, doc.id)) {
+          rememberObservationDoc(doc);
+          inserted++;
+          continue;
+        }
+        try {
+          await insert2(database, doc);
+          rememberObservationDoc(doc);
+          inserted++;
+        } catch {
+        }
+      }
     }
+    await new Promise((resolve2) => setImmediate(resolve2));
   }
   deferredCachedVectorHydration = options.skipCachedVectors ? null : deferCachedVectors ? attachCachedVectors(database, candidates).catch(() => {
   }) : null;
@@ -167080,7 +167097,7 @@ The path should point to a directory containing a .git folder.`
   };
   const server = existingServer ?? new McpServer({
     name: "memorix",
-    version: true ? "1.7.1" : "1.0.1"
+    version: true ? "1.7.2" : "1.0.1"
   });
   const originalRegisterTool = server.registerTool.bind(server);
   server.registerTool = ((name, ...args) => {

package/dist/vector-backfill-runner.js

@@ -5597,7 +5597,7 @@ __export(orama_store_exports, {
   searchObservations: () => searchObservations,
   updateObservationMetadata: () => updateObservationMetadata
 });
-import { create, insert as insert2, search, remove as remove2, update, count, getByID } from "@orama/orama";
+import { create, insert as insert2, insertMultiple, search, remove as remove2, update, count, getByID } from "@orama/orama";
 function getLastSearchMode(projectId) {
   return lastSearchModeByProject.get(projectId ?? SEARCH_MODE_DEFAULT_KEY) ?? "fulltext";
 }
@@ -5809,7 +5809,7 @@ async function hydrateIndex(observations2, options = {}) {
   }
   const deferCachedVectors = options.deferCachedVectors === true && Boolean(indexEmbeddingProvider?.getCachedEmbeddings);
   const cachedVectors = deferCachedVectors ? candidates.map(() => null) : await getCachedEmbeddings(candidates.map(({ observation }) => observationEmbeddingText(observation)));
-  let inserted = 0;
+  const documents = [];
   for (let index = 0; index < candidates.length; index++) {
     const { observation: obs, id } = candidates[index];
     try {
@@ -5843,17 +5843,34 @@ async function hydrateIndex(observations2, options = {}) {
         knowledgeLayer: resolveKnowledgeLayer("observation", obs.sourceDetail, obs.source),
         ...compatibleVector ? { embedding: compatibleVector } : {}
       };
-      await insert2(database, doc);
-      rememberObservationDoc(doc);
-      inserted++;
-      if (inserted % 50 === 0) {
-        await new Promise((resolve) => setImmediate(resolve));
-      }
-      if (inserted % 200 === 0) {
-        await new Promise((resolve) => setTimeout(resolve, 0));
-      }
+      documents.push(doc);
+    } catch {
+    }
+  }
+  let inserted = 0;
+  const hydrationBatchSize = 200;
+  for (let start = 0; start < documents.length; start += hydrationBatchSize) {
+    const batch = documents.slice(start, start + hydrationBatchSize);
+    try {
+      await insertMultiple(database, batch, batch.length);
+      for (const doc of batch) rememberObservationDoc(doc);
+      inserted += batch.length;
     } catch {
+      for (const doc of batch) {
+        if (getByID(database, doc.id)) {
+          rememberObservationDoc(doc);
+          inserted++;
+          continue;
+        }
+        try {
+          await insert2(database, doc);
+          rememberObservationDoc(doc);
+          inserted++;
+        } catch {
+        }
+      }
     }
+    await new Promise((resolve) => setImmediate(resolve));
   }
   deferredCachedVectorHydration = options.skipCachedVectors ? null : deferCachedVectors ? attachCachedVectors(database, candidates).catch(() => {
   }) : null;

package/docs/DEVELOPMENT.md

@@ -13,9 +13,10 @@ Memorix is a TypeScript project built around:
 
 ## Current Development Baseline
 
-The current release work targets the **1.4 line** while package metadata may still show the last published version until the release commit is cut.
+The current release work targets the **1.7 line** while package metadata may
+still show the last published version until the release commit is cut.
 
-Contributors should assume the following areas are part of the 1.4 release line:
+Contributors should assume the following areas are part of the current release line:
 
 - shared memory across MCP clients, CLI, SDK, dashboard, hooks, and memcode
 - memcode uses Memorix project memory, hooks, `/memory` commands, resumable sessions, and model switching as the bundled terminal agent
@@ -27,6 +28,9 @@ Contributors should assume the following areas are part of the 1.4 release line:
 - dashboard config loading aligned with CLI/TUI status surfaces
 - the existing layered retrieval, retention, attribution, and compact output model
 - source-backed long-term memory with candidate, qualification, approval, archival, supersession, and explicit portable user scope
+- preview-first project maintenance shared across Dashboard, CLI, and MCP
+- the reproducible large-store gate for SDK, HTTP MCP, hooks, persistence, and
+  cold-start retrieval budgets
 
 ---
 

package/docs/PERFORMANCE.md

@@ -103,9 +103,16 @@ On the release development machine used for this check, the healthy HTTP service
 temporary Git project with an isolated data directory. It runs with embeddings
 disabled, so it does not use provider credentials or make paid network calls.
 The gate measures steady-state writes, first lexical search, process memory,
-HTTP MCP and hook persistence, and an SDK close/reopen cycle. It also verifies
-that the personal hook candidate remains durable on disk without becoming
-visible to an unbound SDK reader.
+HTTP readiness, MCP initialize/bind/context/search/store, hook capture, and an
+SDK close/reopen cycle. Every network request has a hard timeout, and the
+report fails when an interactive path exceeds its release budget. It also
+verifies that the personal hook candidate remains durable on disk without
+becoming visible to an unbound SDK reader.
+
+Startup hydration uses bounded Orama batches. This keeps HTTP health responsive
+while avoiding a full index rebalance for every persisted observation. The MCP
+search number is intentionally a cold control-plane measurement: it includes
+any lexical hydration the fresh service still owes before its first search.
 
 Results are machine-specific and should be compared on the same OS and Node
 version. The release gate is an integrity and regression check, not a public

package/docs/README.md

@@ -119,11 +119,13 @@ Historical/deep-reference documents may describe older designs. If they conflict
 
 ## Current Product Line
 
-The released product is on the **1.6 line**. The current baseline has:
+The released product is on the **1.7 line**. The current baseline has:
 
 - `memorix setup --agent <agent> --global` is the default agent integration command
 - `memorix serve` is the manual stdio MCP server for external agents
 - `memorix background start` runs the shared HTTP MCP service and dashboard
+- Dashboard maintenance is preview-first, project-scoped, and uses the same
+  cleanup and deduplication planners as CLI and MCP
 - context and CodeGraph commands return bounded task receipts before detailed diagnostics
 - HTTP MCP sessions have a bounded, explicit lifecycle and fail with a reinitialize hint after expiry
 - `memorix integrate --agent <agent>` and `memorix hooks install --agent <agent>` remain manual/fallback generation commands

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "memorix",
   "mcpName": "io.github.AVIDS2/memorix",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "description": "Local-first shared memory layer for AI coding agents across MCP clients, Git history, reasoning context, and project sessions.",
   "type": "module",
   "sideEffects": false,

package/plugins/claude/memorix/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://anthropic.com/claude-code/plugin.schema.json",
   "name": "memorix",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "description": "Shared workspace memory for Claude Code and other AI coding agents.",
   "author": {
     "name": "AVIDS2",

package/plugins/codebuddy/memorix/.codebuddy-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "memorix",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "description": "Local-first project memory for CodeBuddy Code and other coding agents.",
   "author": {
     "name": "AVIDS2",

package/plugins/codex/memorix/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "memorix",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "description": "Shared workspace memory for Codex and other AI coding agents.",
   "author": {
     "name": "AVIDS2",

package/plugins/copilot/memorix/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "memorix",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "description": "Shared local workspace memory for GitHub Copilot CLI and other AI coding agents.",
   "author": {
     "name": "AVIDS2",

package/plugins/gemini/memorix/gemini-extension.json

@@ -1,6 +1,6 @@
 {
   "name": "memorix",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "mcpServers": {
     "memorix": {
       "command": "memorix",

package/plugins/omp/memorix/package.json

@@ -1,6 +1,6 @@
 {
   "name": "memorix-omp-package",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "description": "Memorix shared workspace memory package for Oh-my-Pi.",
   "license": "Apache-2.0",
   "keywords": [

package/plugins/openclaw/memorix/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "memorix",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "description": "Shared workspace memory for OpenClaw and other AI coding agents.",
   "author": {
     "name": "AVIDS2",

package/plugins/pi/memorix/package.json

@@ -1,6 +1,6 @@
 {
   "name": "memorix-pi-package",
-  "version": "1.7.1",
+  "version": "1.7.2",
   "description": "Memorix shared workspace memory package for Pi coding agent.",
   "license": "Apache-2.0",
   "keywords": [

package/server.json

@@ -7,7 +7,7 @@
     "url": "https://github.com/AVIDS2/memorix",
     "source": "github"
   },
-  "version": "1.7.1",
+  "version": "1.7.2",
   "websiteUrl": "https://github.com/AVIDS2/memorix#readme",
   "icons": [
     {
@@ -22,7 +22,7 @@
     {
       "registryType": "npm",
       "identifier": "memorix",
-      "version": "1.7.1",
+      "version": "1.7.2",
       "runtimeHint": "npx",
       "packageArguments": [
         {

package/src/store/orama-store.ts

@@ -8,7 +8,7 @@
  * Vector search (embeddings) will be added in P1 phase.
  */
 
-import { create, insert, search, remove, update, count, getByID, type AnyOrama } from '@orama/orama';
+import { create, insert, insertMultiple, search, remove, update, count, getByID, type AnyOrama } from '@orama/orama';
 import type { MemorixDocument, SearchOptions, IndexEntry, KnowledgeLayer, ObservationReader } from '../types.js';
 import { OBSERVATION_ICONS, type ObservationType } from '../types.js';
 import { resolveKnowledgeLayer } from '../skills/mini-skills.js';
@@ -405,7 +405,7 @@ export async function hydrateIndex(
     ? candidates.map(() => null)
     : await getCachedEmbeddings(candidates.map(({ observation }) => observationEmbeddingText(observation)));
 
-  let inserted = 0;
+  const documents: MemorixDocument[] = [];
   for (let index = 0; index < candidates.length; index++) {
     const { observation: obs, id } = candidates[index];
     try {
@@ -439,19 +439,39 @@ export async function hydrateIndex(
         knowledgeLayer: resolveKnowledgeLayer('observation', obs.sourceDetail, obs.source),
         ...(compatibleVector ? { embedding: compatibleVector } : {}),
       };
-      await insert(database, doc);
-      rememberObservationDoc(doc);
-      inserted++;
-      // Keep the HTTP event loop alive on large corpora so /health still answers.
-      if (inserted % 50 === 0) {
-        await new Promise<void>((resolve) => setImmediate(resolve));
-      }
-      if (inserted % 200 === 0) {
-        await new Promise<void>((resolve) => setTimeout(resolve, 0));
-      }
+      documents.push(doc);
     } catch { /* skip malformed entries */ }
   }
 
+  let inserted = 0;
+  const hydrationBatchSize = 200;
+  for (let start = 0; start < documents.length; start += hydrationBatchSize) {
+    const batch = documents.slice(start, start + hydrationBatchSize);
+    try {
+      // Orama can rebalance its indexes once per batch instead of once per
+      // document. Keep batches bounded so HTTP health remains responsive.
+      await insertMultiple(database, batch, batch.length);
+      for (const doc of batch) rememberObservationDoc(doc);
+      inserted += batch.length;
+    } catch {
+      // A concurrent write or malformed document can partially complete a
+      // batch. Preserve the old best-effort behavior without double-inserting.
+      for (const doc of batch) {
+        if (getByID(database, doc.id)) {
+          rememberObservationDoc(doc);
+          inserted++;
+          continue;
+        }
+        try {
+          await insert(database, doc);
+          rememberObservationDoc(doc);
+          inserted++;
+        } catch { /* skip malformed or concurrently inserted entries */ }
+      }
+    }
+    await new Promise<void>((resolve) => setImmediate(resolve));
+  }
+
   deferredCachedVectorHydration = options.skipCachedVectors
     ? null
     : deferCachedVectors