recallnest 2.2.0 → 2.2.1

package/.env

@@ -0,0 +1,4 @@
+JINA_API_KEY=jina_a953576fa7b746019e57d1370122fb8aNQLVOjq95WHeDBKlBr3JeCj0K5SL
+QWEN_API_KEY=sk-8b8f215de882422a991653d5fc3cc275
+RECALLNEST_EMOTION_SCORING=true
+RECALLNEST_CONSTRUCTIVE_RETRIEVAL=true

package/config.json

@@ -0,0 +1,55 @@
+{
+  "dbPath": "./data/lancedb",
+  "recallMode": "full",
+  "embedding": {
+    "provider": "jina",
+    "apiKey": "${JINA_API_KEY}",
+    "model": "jina-embeddings-v5-text-small",
+    "baseURL": "https://api.jina.ai/v1",
+    "dimensions": 1024,
+    "taskQuery": "retrieval.query",
+    "taskPassage": "retrieval.passage"
+  },
+  "llm": {
+    "apiKey": "${QWEN_API_KEY}",
+    "model": "qwen-turbo",
+    "baseURL": "https://dashscope.aliyuncs.com/compatible-mode/v1"
+  },
+  "sources": {
+    "cc": {
+      "path": "auto",
+      "glob": "*.jsonl",
+      "description": "Claude Code transcripts"
+    },
+    "codex": {
+      "path": "~/.codex/sessions",
+      "glob": "*.jsonl",
+      "description": "Codex sessions"
+    },
+    "memory": {
+      "path": "auto",
+      "glob": "*.md",
+      "description": "Claude Code memory files"
+    },
+    "gemini": {
+      "path": "~/.gemini/tmp",
+      "glob": "session-*.json",
+      "description": "Gemini CLI sessions"
+    },
+    "desktop": {
+      "path": "./data/desktop-import",
+      "glob": "*.jsonl",
+      "description": "Claude Desktop / claude.ai web conversations"
+    }
+  },
+  "retrieval": {
+    "mode": "hybrid",
+    "vectorWeight": 0.6,
+    "bm25Weight": 0.4,
+    "recencyHalfLifeDays": 30,
+    "timeDecayHalfLifeDays": 60,
+    "hardMinScore": 0.38,
+    "hotnessWeight": 0.15
+  },
+  "recallDepthDefault": "l1"
+}

package/logs/daily-recall.log

@@ -0,0 +1,30 @@
+=== 2026-04-12 08:00:01 daily-recall 开始 ===
+=== 2026-04-12 08:00:01 daily-recall 开始 ===
+  今日主题 (#7): 踩坑 修复 经验总结
+  今日主题 (#7): 踩坑 修复 经验总结
+  检查 distill gate...
+  检查 distill gate...
+  distill gate: 
+  distill gate: 
+  执行 distill...
+  执行 distill...
+  distill 完成，27 行
+  distill 完成，27 行
+  [完成] 回顾已写入: /Users/anxianjingya/recallnest/daily-reflections/2026-04-12.md
+  [完成] 回顾已写入: /Users/anxianjingya/recallnest/daily-reflections/2026-04-12.md
+  执行 checkpoint-gc...
+  执行 checkpoint-gc...
+  checkpoint-gc: === Checkpoint GC ===
+
+Kept     : 217
+Archived : 0
+Deleted  : 0
+Errors   : 0
+  checkpoint-gc: === Checkpoint GC ===
+
+Kept     : 217
+Archived : 0
+Deleted  : 0
+Errors   : 0
+=== 2026-04-12 08:00:13 daily-recall 结束 ===
+=== 2026-04-12 08:00:13 daily-recall 结束 ===

package/logs/ingest-2026-04-11.log

@@ -0,0 +1,110 @@
+=== 2026-04-11 19:22:04 增量更新开始 ===
+[INFO] LLM client initialized: qwen-turbo @ https://dashscope.aliyuncs.com/compatible-mode/v1
+
+🔑 验证 Embedding API...
+  ✅ jina-embeddings-v5-text-small (1024d)
+  ✅ LLM: qwen-turbo (L0 摘要 + 语义去重)
+  ✅ 两阶段去重: vector + LLM 语义判断
+
+🔄 开始导入记忆 (source: all)...
+
+📝 导入 Claude Code 对话...
+  CC: 230/962 files, 120 chunks ingested
+  CC: 470/962 files, 186 chunks ingested
+[INFO] noise-filter: denial pattern matched: "**Phase 5: Predictive Prospective Memory — DONE** ✅
+
+总结一下做了什..."
+  ✅ CC: 24 files, 528 ingested, 474 deduped (hard:17, exact:160, llm-skip:150, llm-merge:147), 0 errors
+🤖 导入 Codex 对话...
+[INFO] noise-filter: meta-question filtered: "# AGENTS.md instructions for /Users/anxianjingya/Projects/re..."
+[INFO] noise-filter: meta-question filtered: "# AGENTS.md instructions for /Users/anxianjingya
+
+<INSTRUCTI..."
+  Codex: 294/294 files, 39 chunks ingested
+  ✅ Codex: 2 files, 39 ingested, 22 deduped (hard:6, exact:0, llm-skip:8, llm-merge:8), 0 errors
+💎 导入 Gemini 对话...
+  ✅ Gemini: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+🖥️  导入 Claude Desktop 对话...
+  CC: 10/349 files, 43 chunks ingested
+[INFO] noise-filter: denial pattern matched: "好的！我来手把手教你！别担心，超简单！ 😊
+
+---
+
+## 📍 **Step 1: 找到文件在哪**
+
+我刚才给你..."
+  CC: 20/349 files, 70 chunks ingested
+  CC: 30/349 files, 73 chunks ingested
+[INFO] noise-filter: short boilerplate filtered: "好的，我来帮你截图。"
+  CC: 40/349 files, 160 chunks ingested
+  CC: 50/349 files, 180 chunks ingested
+  CC: 60/349 files, 226 chunks ingested
+  CC: 70/349 files, 256 chunks ingested
+[INFO] noise-filter: meta-question filtered: "你记不记得我让你帮我弄孩子们的简历..."
+[INFO] noise-filter: denial pattern matched: "我查找一下我们之前关于孩子们简历的对话...
+
+让我搜索一下相关内容。
+
+我没有找到我们之前讨论过帮孩子们制作简历的记录..."
+  CC: 80/349 files, 316 chunks ingested
+  CC: 90/349 files, 335 chunks ingested
+  CC: 100/349 files, 412 chunks ingested
+  CC: 110/349 files, 501 chunks ingested
+  CC: 120/349 files, 520 chunks ingested
+[INFO] noise-filter: boilerplate filtered: "Hello! How can I help you today?..."
+[INFO] noise-filter: denial pattern matched: "不能。我无法访问你本地电脑上的文件或 Claude Code 的 session 记录。
+
+我能访问的是：
+1. **我..."
+[INFO] noise-filter: denial pattern matched: "这个我不太确定，因为 Claude Desktop 的会话索引存在 `~/Library/Application Sup..."
+[INFO] noise-filter: boilerplate filtered: "Hi Alice,
+I understand you're looking for a way to rebuild t..."
+  CC: 130/349 files, 547 chunks ingested
+  CC: 140/349 files, 585 chunks ingested
+[INFO] noise-filter: denial pattern matched: "啊，你说得对！我理解错了。
+
+让我重新想想这个问题...
+
+Antigravity 确实也是通过 Claude in C..."
+  CC: 150/349 files, 618 chunks ingested
+[INFO] noise-filter: denial pattern matched: "我来帮你查看workspace里的PPT相关文件，分析其逻辑是否与截图中描述的一致。
+
+让我先看看目录结构：
+我看到wo..."
+  CC: 160/349 files, 659 chunks ingested
+  CC: 170/349 files, 681 chunks ingested
+[INFO] noise-filter: denial pattern matched: "我不确定Claude Opus 4.5的具体知识截止日期是否与我的相同。
+
+让我搜索一下Anthropic官方文档来确认..."
+  CC: 180/349 files, 685 chunks ingested
+  CC: 190/349 files, 694 chunks ingested
+  CC: 210/349 files, 919 chunks ingested
+  CC: 220/349 files, 927 chunks ingested
+  CC: 230/349 files, 956 chunks ingested
+  CC: 240/349 files, 977 chunks ingested
+  CC: 250/349 files, 1018 chunks ingested
+  CC: 260/349 files, 1030 chunks ingested
+  CC: 270/349 files, 1048 chunks ingested
+  CC: 280/349 files, 1185 chunks ingested
+  CC: 290/349 files, 1196 chunks ingested
+[INFO] noise-filter: boilerplate filtered: "Hi Alice! Here's what I've learned about you from our past c..."
+  CC: 300/349 files, 1211 chunks ingested
+  CC: 310/349 files, 1243 chunks ingested
+  CC: 320/349 files, 1276 chunks ingested
+[INFO] noise-filter: denial pattern matched: "我想不起来给你的这些提示词和视觉skill是哪里来的..."
+[INFO] noise-filter: denial pattern matched: "* 全网AI集体翻车？为了读一篇公众号文章，我扒开了Grok的“谎言”
+ * Gemini、Claude、元宝全试了，竟..."
+[INFO] noise-filter: denial pattern matched: "让我先查看歌舞剧院的总结，提取人才相关数据：
+好的,我现在已经了解了录音中领导的修改意见和两家院团的总结内容。根据领导的..."
+  CC: 330/349 files, 1331 chunks ingested
+  CC: 340/349 files, 1348 chunks ingested
+  CC: 349/349 files, 1372 chunks ingested
+  ✅ Desktop: 343 files, 1372 ingested, 3805 deduped (hard:440, exact:1721, llm-skip:1045, llm-merge:599), 0 errors
+📚 导入记忆文件...
+  ✅ Memory: 8 files, 11 ingested, 279 deduped (hard:0, exact:257, llm-skip:15, llm-merge:7), 0 errors
+
+📊 导入汇总:
+  总计: 1950 chunks 已索引, 4580 chunks 去重 (hard:463, exact:2138, llm-skip:1218, llm-merge:761), 0 errors
+
+=== 2026-04-11 20:37:10 增量更新完成 ===
+

package/logs/ingest-2026-04-12.log

@@ -0,0 +1,90 @@
+=== 2026-04-12 03:00:01 增量更新开始 ===
+[INFO] LLM client initialized: qwen-turbo @ https://dashscope.aliyuncs.com/compatible-mode/v1
+
+🔑 验证 Embedding API...
+  ✅ jina-embeddings-v5-text-small (1024d)
+  ✅ LLM: qwen-turbo (L0 摘要 + 语义去重)
+  ✅ 两阶段去重: vector + LLM 语义判断
+
+🔄 开始导入记忆 (source: all)...
+
+📝 导入 Claude Code 对话...
+[INFO] noise-filter: denial pattern matched: "<task-notification>
+<task-id>ac1e8c3a46f90bcc6</task-id>
+<to..."
+  CC: 400/976 files, 269 chunks ingested
+[INFO] noise-filter: denial pattern matched: "完全破案了。
+
+**根因：4月10日 22:55（新加坡时间）发生了一次批量仓库可见性修改**
+
+时间线还原：
+
+| U..."
+  CC: 770/976 files, 541 chunks ingested
+  ✅ CC: 19 files, 563 ingested, 505 deduped (hard:24, exact:228, llm-skip:114, llm-merge:139), 0 errors
+🤖 导入 Codex 对话...
+  ✅ Codex: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+💎 导入 Gemini 对话...
+  ✅ Gemini: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+🖥️  导入 Claude Desktop 对话...
+  ✅ Desktop: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+📚 导入记忆文件...
+  ✅ Memory: 5 files, 4 ingested, 188 deduped (hard:0, exact:172, llm-skip:13, llm-merge:3), 0 errors
+
+📊 导入汇总:
+  总计: 567 chunks 已索引, 693 chunks 去重 (hard:24, exact:400, llm-skip:127, llm-merge:142), 0 errors
+
+=== 2026-04-12 03:16:02 增量更新完成 ===
+
+=== 2026-04-12 11:16:02 增量更新开始 ===
+[INFO] LLM client initialized: qwen-turbo @ https://dashscope.aliyuncs.com/compatible-mode/v1
+
+🔑 验证 Embedding API...
+  ✅ jina-embeddings-v5-text-small (1024d)
+  ✅ LLM: qwen-turbo (L0 摘要 + 语义去重)
+  ✅ 两阶段去重: vector + LLM 语义判断
+
+🔄 开始导入记忆 (source: all)...
+
+📝 导入 Claude Code 对话...
+  ✅ CC: 5 files, 141 ingested, 90 deduped (hard:4, exact:24, llm-skip:30, llm-merge:32), 0 errors
+🤖 导入 Codex 对话...
+  ✅ Codex: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+💎 导入 Gemini 对话...
+  ✅ Gemini: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+🖥️  导入 Claude Desktop 对话...
+  ✅ Desktop: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+📚 导入记忆文件...
+  ✅ Memory: 1 files, 0 ingested, 182 deduped (hard:0, exact:171, llm-skip:11, llm-merge:0), 0 errors
+
+📊 导入汇总:
+  总计: 141 chunks 已索引, 272 chunks 去重 (hard:4, exact:195, llm-skip:41, llm-merge:32), 0 errors
+
+=== 2026-04-12 11:21:02 增量更新完成 ===
+
+=== 2026-04-12 19:21:02 增量更新开始 ===
+[INFO] LLM client initialized: qwen-turbo @ https://dashscope.aliyuncs.com/compatible-mode/v1
+
+🔑 验证 Embedding API...
+  ✅ jina-embeddings-v5-text-small (1024d)
+  ✅ LLM: qwen-turbo (L0 摘要 + 语义去重)
+  ✅ 两阶段去重: vector + LLM 语义判断
+
+🔄 开始导入记忆 (source: all)...
+
+📝 导入 Claude Code 对话...
+  ✅ CC: 3 files, 55 ingested, 188 deduped (hard:2, exact:134, llm-skip:33, llm-merge:19), 0 errors
+🤖 导入 Codex 对话...
+  ✅ Codex: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+💎 导入 Gemini 对话...
+  ✅ Gemini: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+🖥️  导入 Claude Desktop 对话...
+  ✅ Desktop: 0 files, 0 ingested, 0 deduped (hard:0, exact:0, llm-skip:0, llm-merge:0), 0 errors
+📚 导入记忆文件...
+  ✅ Memory: 1 files, 0 ingested, 182 deduped (hard:0, exact:171, llm-skip:11, llm-merge:0), 0 errors
+
+📊 导入汇总:
+  总计: 55 chunks 已索引, 370 chunks 去重 (hard:2, exact:305, llm-skip:44, llm-merge:19), 0 errors
+
+=== 2026-04-12 19:25:03 增量更新完成 ===
+

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "recallnest",
-  "version": "2.2.0",
+  "version": "2.2.1",
   "description": "RecallNest \u2014 MCP-native, local-first memory search and distillation for AI conversations",
   "license": "MIT",
   "type": "module",

package/.claude-plugin/CLAUDE.md

@@ -1,29 +0,0 @@
-# RecallNest Memory Plugin
-
-RecallNest is a local-first shared memory layer backed by LanceDB. It stores and recalls context across sessions, terminals, and agents (Claude Code, Codex, Gemini CLI).
-
-**Prerequisite**: Requires [Bun](https://bun.sh) runtime. On first start, dependencies install automatically.
-
-## Tool Quick Reference
-
-| When | Tool | Purpose |
-|------|------|---------|
-| Session opens, user says continue/接着/上个窗口 | `resume_context` | Load stable context, latest checkpoint, recent cases |
-| Before answering about prior work / known facts | `search_memory` | Targeted fact/pattern recall (2-3 key nouns) |
-| After solving something significant | `store_memory` or `store_case` | Persist durable knowledge |
-| Before closing a window with unfinished work | `checkpoint_session` | Save current state for next session |
-| After discovering a reusable workflow | `store_workflow_pattern` | Persist as durable pattern |
-| Memory feels stale or conflicted | `consolidate_memories` | Merge near-duplicates (dry-run by default) |
-
-## Continuity Rules
-
-- If the user says `continue`, `继续`, `接着`, `刚才`, `上个窗口`, `不要让我重复前情`, or asks where you stopped / what to do next — call `resume_context` before any Bash, Read, or repo exploration.
-- If the user starts a concrete task in an active project or asks for prior implementation detail — do lightweight RecallNest recovery before repo exploration even without an explicit `continue`.
-- Do not substitute `git status`, `git log`, or reading local docs for `resume_context`. Local inspection can validate current code state only **after** continuity has been recovered.
-- When `scope` or `sessionId` is known, pass it into recall tools instead of relying on global recall.
-- If `resume_context` returns a resolved scope or checkpoint scope, reuse that same scope in follow-up `search_memory`, `brief_memory`, `pin_memory` calls.
-- If recalled context mentions repo state (modified files, pending push), treat it as **handoff context only** — do not restate as current state without verifying in this window.
-- Before leaving a window with resumable work, call `checkpoint_session`. Do not include repo state claims unless you actually inspected the repo this window.
-- Use `store_memory` for durable profile, preference, entity, or case knowledge.
-- Use `store_workflow_pattern` for reusable multi-step workflows.
-- If the user explicitly corrects a continuity miss, fix the workflow and call `workflow_observe` with `outcome: corrected`.

package/.claude-plugin/marketplace.json

@@ -1,18 +0,0 @@
-{
-  "name": "AliceLJY",
-  "owner": {
-    "name": "Alice Jin Ye"
-  },
-  "metadata": {
-    "description": "RecallNest — shared memory layer for Claude Code, Codex, and Gemini CLI",
-    "homepage": "https://github.com/AliceLJY/recallnest"
-  },
-  "plugins": [
-    {
-      "name": "recallnest",
-      "version": "1.4.1",
-      "source": "./",
-      "description": "Local-first memory system: store, search, and recall context across sessions and terminals. Powered by LanceDB with hybrid retrieval."
-    }
-  ]
-}

package/.dockerignore

@@ -1,11 +0,0 @@
-node_modules/
-data/*
-!data/alias-map.example.json
-logs/
-.env
-.git/
-*.md
-src/__tests__/
-eval/
-integrations/
-scripts/

package/.github/workflows/ci.yml

@@ -1,28 +0,0 @@
-name: CI
-
-on:
-  push:
-    branches: [main]
-  pull_request:
-    branches: [main]
-
-jobs:
-  check:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-
-      - uses: oven-sh/setup-bun@v2
-        with:
-          bun-version: latest
-
-      - run: bun install
-
-      - name: Setup config
-        run: cp config.json.example config.json
-
-      - name: Doctor (CI mode)
-        run: bun run src/cli.ts doctor --ci
-
-      - name: TypeScript check
-        run: bun build src/cli.ts --target bun --no-bundle --outdir /tmp/check 2>&1 || true

package/.mcp.json

@@ -1,9 +0,0 @@
-{
-  "mcpServers": {
-    "recallnest": {
-      "type": "stdio",
-      "command": "bash",
-      "args": ["${CLAUDE_PLUGIN_ROOT}/scripts/start-server.sh"]
-    }
-  }
-}

package/CHANGELOG.md

@@ -1,51 +0,0 @@
-# Changelog
-
-## v1.3.1 — Upstream Sync (2026-03-12)
-
-Synced with [CortexReach/memory-lancedb-pro](https://github.com/CortexReach/memory-lancedb-pro) master (v1.1.0-beta.6+).
-
-### Changed
-
-- **Retriever**: Added `source` field to `RetrievalContext` — access reinforcement now only fires on manual retrieval, preventing auto-recall from strengthening noise memories (synced from upstream beta.2 design).
-- **Noise filter**: Added Chinese meta-question patterns (`你记得`, `记不记得`, `还记得…吗`, `上次…说`, `之前…提到`) and diagnostic artifact filter (synced from upstream beta.3).
-- **README**: Updated upstream credit link from `win4r/memory-lancedb-pro` to `CortexReach/memory-lancedb-pro`, added CortexReach team acknowledgement.
-
-## v1.2.0 — First Distributable Release (2026-03-08)
-
-The goal of this release: a new user can go from `git clone` to first search result in 15 minutes.
-
-### New
-
-- **`lm doctor`** — one-command pre-flight check for Bun, config, API key, data directory, transcript paths, and index health. Supports `--ci` mode for GitHub Actions.
-- **`lm demo`** — run sample queries to see RecallNest in action before writing your own.
-- **`config.json.example`** — ships with absolute `~/.recallnest/data/lancedb` path. New users copy this instead of editing the tracked config.
-- **GitHub Actions CI** — runs `doctor --ci` and TypeScript check on every push.
-- **Ingest pre-validation** — embedding API is tested before processing any files. Invalid Jina key now fails fast with a clear message instead of crashing mid-ingest.
-
-### Changed
-
-- **README rewritten** — added Prerequisites table (Bun + Jina key), 5-step quickstart with expected output, Troubleshooting section.
-- **Gemini support marked "coming soon"** — README, config example, and doctor all honestly reflect that Gemini CLI sessions are encrypted protobuf and not yet parseable. The `lm ingest` command prints a clear skip message instead of silently failing.
-- **Config path robustness** — default `dbPath` changed from relative `./data/lancedb` to absolute `~/.recallnest/data/lancedb` in config example. Auto-detect failure messages now include the user's actual home path.
-- **`config.json` untracked** — added to `.gitignore` so user config is not overwritten by `git pull`.
-
-### Fixed
-
-- Auto-detect hint in `doctor` now shows a real example path based on the current user's home directory.
-- `findConfigPath()` error message now suggests `cp config.json.example config.json` when the example file exists.
-
-## v1.1.0 — Hybrid Retrieval + MCP + UI (2026-02)
-
-- Hybrid retrieval: LanceDB vector + BM25 keyword search with configurable weights
-- Retrieval profiles: `default`, `writing`, `debug`, `fact-check`
-- MCP server with 9 tools: search, explain, distill, brief, pin, list assets/pins, export, stats
-- Local web workbench UI at `http://localhost:4317`
-- Multi-source ingest: Claude Code transcripts, Codex sessions, Gemini sessions, markdown notes
-- Asset system: pin, brief, export with re-indexing
-- Time-aware scoring with configurable decay
-
-## v1.0.0 — Initial Release (2026-01)
-
-- Basic vector search over Claude Code transcripts
-- LanceDB storage with Jina embeddings
-- CLI interface

package/DEMO.md

@@ -1,94 +0,0 @@
-# RecallNest 30-Second Demo
-
-Use this script when you want to explain RecallNest fast on GitHub, X, Telegram, or in a live screen recording.
-
-## Demo Goal
-
-Show three things in one pass:
-
-| What to prove | What the viewer should understand |
-|------|------|
-| Recall | old AI conversations can be found quickly |
-| Explainability | hits are not opaque |
-| Reusability | useful context can become a long-lived asset |
-
-## Fast Setup
-
-```bash
-cd /Users/anxianjingya/recallnest
-bun run src/ui-server.ts
-```
-
-Open:
-
-```text
-http://localhost:4317
-```
-
-## Recommended Demo Query
-
-Use:
-
-```text
-telegram bridge
-```
-
-Reason:
-
-| Why this query works | Result |
-|------|------|
-| already appears in the current dataset | likely to return hits immediately |
-| has operational context | good for showing explainability |
-| already has assets | good for showing reuse |
-
-## 30-Second Talk Track
-
-### Version A
-
-```text
-This is RecallNest, a local-first memory workbench for AI conversations.
-It turns Claude Code, Codex, Gemini, and notes into a recall layer.
-I can search an old topic, inspect why it matched, then pin useful context back into memory so it can be recalled again later.
-The same memory layer is exposed through MCP, so an agent can use it and I can still verify it in the UI.
-```
-
-### Version B
-
-```text
-RecallNest is not just transcript search.
-It lets me search, explain, distill, and reuse past AI conversations.
-The key idea is that useful hits do not disappear again: I can turn them into assets and feed them back into recall.
-```
-
-## Demo Click Path
-
-| Step | UI action | What to say |
-|------|------|------|
-| 1 | show the loaded `telegram bridge` query | "This starts from a real transcript topic." |
-| 2 | click `Search` | "I can recall the relevant conversation history." |
-| 3 | click `Details` on one result | "The hit is inspectable, not a black box." |
-| 4 | click `Pin` or `Brief` | "Useful context becomes a reusable asset." |
-| 5 | switch to `Assets` | "Assets stay visible and can be recalled later." |
-
-## Screenshot Rules
-
-| Rule | Why |
-|------|------|
-| keep the browser crop tight | the product, not Chrome, should dominate the frame |
-| show real hits in the result area | empty states spread poorly |
-| keep `Trace Output` collapsed unless you are explaining it | it distracts from the main product story |
-| prefer one strong screenshot over many average ones | easier to share and understand |
-
-## If The UI Opens Empty
-
-Check:
-
-1. the server is running
-2. the current dataset was ingested
-3. the page has been refreshed after startup
-
-If needed, re-run:
-
-```bash
-bun run src/ui-server.ts
-```

package/Dockerfile

@@ -1,20 +0,0 @@
-FROM oven/bun:1-alpine AS base
-WORKDIR /app
-
-# Install dependencies
-COPY package.json ./
-RUN bun install --frozen-lockfile 2>/dev/null || bun install
-
-# Copy source
-COPY src/ src/
-COPY data/alias-map.example.json data/alias-map.example.json
-
-# Default environment
-ENV RECALLNEST_API_PORT=4318
-ENV RECALLNEST_UI_PORT=4317
-
-EXPOSE 4318 4317
-
-# MCP stdio mode (default) — Glama introspection connects here
-# Override with CMD ["bun","run","src/api-server.ts"] for HTTP mode
-CMD ["bun", "run", "src/mcp-server.ts"]

package/EVAL.md

@@ -1,89 +0,0 @@
-# RecallNest Eval Memo
-
-This file is the operator memo for improving retrieval quality without relying on intuition.
-
-## Why this exists
-
-RecallNest will keep evolving with:
-- upstream retrieval changes from `memory-lancedb-pro`
-- local tuning for your own workflows
-- new asset behavior (`pin`, `brief`, cleanup rules)
-
-Without a repeatable eval, it is impossible to tell whether recall actually improved.
-
-## Core files
-
-| File | Purpose |
-|------|---------|
-| `eval/cases.json` | real-world recall cases to protect |
-| `src/eval.ts` | eval runner |
-| `eval/reports/` | saved baselines to compare across upgrades |
-
-## What to evaluate
-
-Use queries that matter in real usage:
-- bot / bridge maintenance
-- OpenClaw memory architecture
-- writing style and user preference recall
-- visual style preference recall
-- AWS / config operations
-
-Important:
-- prefer the wording the operator actually types
-- do not invent a cleaner label if the real query is messier but more common
-- if `aws ssh` is what gets used in practice, benchmark `aws ssh`, not `aws bot config`
-- abstract, summarized queries are valid benchmarks if that is how the operator naturally searches
-- protect both query styles: exact operational wording and high-level conceptual wording
-
-Each case should define:
-- `query`
-- `profile`
-- optional `scope`
-- `expectAny`
-- `expectAll`
-- `expectScopePrefixes`
-- optional `forbid`
-
-## Recommended workflow
-
-### Before changing retrieval
-
-Run:
-
-```bash
-bun run src/eval.ts --output eval/reports/latest.md
-```
-
-If the change is important, also save a dated snapshot:
-
-```bash
-bun run src/eval.ts --output eval/reports/2026-03-06-baseline.md
-```
-
-### After changing retrieval
-
-Run the same eval again and compare:
-- pass count
-- average score
-- top scopes
-- top snippet quality
-
-## What counts as a good change
-
-Good:
-- relevant source scopes move up
-- user preference memories stay stable
-- bridge / ops / config queries become easier to recover
-- `pin` and `brief` behave differently on purpose
-
-Bad:
-- asset recursion comes back
-- old noisy `brief` objects dominate retrieval
-- exact operational queries stop surfacing recent config changes
-- writing/style queries drift away from user preference memories
-
-## Maintenance note
-
-Whenever a new recurring workflow appears, add it to `eval/cases.json`.
-
-That turns “I hope this still works” into “I can prove it still works.”