recallnest 2.2.0 → 2.2.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (232) hide show
  1. package/.env +4 -0
  2. package/config.json +55 -0
  3. package/logs/daily-recall.error.log +0 -0
  4. package/logs/daily-recall.log +30 -0
  5. package/logs/ingest-2026-04-11.log +110 -0
  6. package/logs/ingest-2026-04-12.log +90 -0
  7. package/logs/launchagent-stderr.log +0 -0
  8. package/logs/launchagent-stdout.log +0 -0
  9. package/package.json +1 -1
  10. package/.claude-plugin/CLAUDE.md +0 -29
  11. package/.claude-plugin/marketplace.json +0 -18
  12. package/.dockerignore +0 -11
  13. package/.github/workflows/ci.yml +0 -28
  14. package/.mcp.json +0 -9
  15. package/CHANGELOG.md +0 -51
  16. package/DEMO.md +0 -94
  17. package/Dockerfile +0 -20
  18. package/EVAL.md +0 -89
  19. package/FAILURES.md +0 -75
  20. package/OPERATIONS.md +0 -59
  21. package/ROADMAP.md +0 -245
  22. package/SKILL.md +0 -250
  23. package/assets/dashboard.png +0 -0
  24. package/assets/knowledge-graph.png +0 -0
  25. package/assets/recallnest-workbench.png +0 -0
  26. package/assets/screenshots/ui-full.png +0 -0
  27. package/assets/screenshots/ui-overview.png +0 -0
  28. package/assets/screenshots/ui-search.png +0 -0
  29. package/assets/ui/app.js +0 -1031
  30. package/assets/ui/index.html +0 -224
  31. package/assets/ui/styles.css +0 -781
  32. package/assets/wechat-qr.jpg +0 -0
  33. package/bun.lock +0 -337
  34. package/daily-reflections/2026-03-05.md +0 -15
  35. package/daily-reflections/2026-03-06.md +0 -15
  36. package/daily-reflections/2026-03-07.md +0 -7
  37. package/daily-reflections/2026-03-08.md +0 -34
  38. package/daily-reflections/2026-03-09.md +0 -35
  39. package/daily-reflections/2026-03-10.md +0 -36
  40. package/daily-reflections/2026-03-11.md +0 -35
  41. package/daily-reflections/2026-03-12.md +0 -36
  42. package/daily-reflections/2026-03-13.md +0 -35
  43. package/daily-reflections/2026-03-14.md +0 -35
  44. package/daily-reflections/2026-03-15.md +0 -35
  45. package/daily-reflections/2026-03-16.md +0 -41
  46. package/daily-reflections/2026-03-17.md +0 -10
  47. package/daily-reflections/2026-03-18.md +0 -10
  48. package/daily-reflections/2026-03-19.md +0 -10
  49. package/daily-reflections/2026-03-20.md +0 -10
  50. package/daily-reflections/2026-03-21.md +0 -10
  51. package/daily-reflections/2026-03-22.md +0 -10
  52. package/daily-reflections/2026-03-23.md +0 -10
  53. package/daily-reflections/2026-03-24.md +0 -10
  54. package/daily-reflections/2026-03-25.md +0 -32
  55. package/daily-reflections/2026-03-26.md +0 -37
  56. package/daily-reflections/2026-03-27.md +0 -35
  57. package/daily-reflections/2026-03-28.md +0 -35
  58. package/daily-reflections/2026-03-29.md +0 -35
  59. package/daily-reflections/2026-03-30.md +0 -36
  60. package/daily-reflections/2026-03-31.md +0 -35
  61. package/daily-reflections/2026-04-01.md +0 -35
  62. package/daily-reflections/2026-04-02.md +0 -35
  63. package/daily-reflections/2026-04-03.md +0 -37
  64. package/daily-reflections/2026-04-04.md +0 -37
  65. package/daily-reflections/2026-04-05.md +0 -35
  66. package/daily-reflections/2026-04-06.md +0 -36
  67. package/daily-reflections/2026-04-07.md +0 -35
  68. package/daily-reflections/2026-04-08.md +0 -35
  69. package/daily-reflections/2026-04-12.md +0 -35
  70. package/data/alias-map.example.json +0 -22
  71. package/docs/api-reference.md +0 -1291
  72. package/docs/architecture.md +0 -168
  73. package/docs/memory-boundary-contract.md +0 -226
  74. package/docs/memory-categories.md +0 -119
  75. package/docs/pipeline-audit-ultramemory.md +0 -177
  76. package/docs/plans/2026-03-08-query-expansion.md +0 -270
  77. package/docs/plans/2026-03-16-continuity-layer.md +0 -196
  78. package/docs/plans/2026-04-01-recall-boost.md +0 -100
  79. package/docs/plans/mcp-tiering-plan.md +0 -126
  80. package/docs/research/recall-quality-audit-2026-03-22.md +0 -153
  81. package/docs/self-evolution.md +0 -198
  82. package/docs/superpowers/plans/2026-03-19-ingest-pipeline-upgrade.md +0 -782
  83. package/docs/superpowers/plans/2026-04-11-constructive-retrieval-emotion.md +0 -1195
  84. package/docs/superpowers/specs/2026-03-19-ingest-pipeline-upgrade-design.md +0 -77
  85. package/docs/superpowers/specs/2026-03-23-pull-claude-desktop-design.md +0 -95
  86. package/docs/superpowers/specs/2026-04-11-constructive-retrieval-emotion-design.md +0 -398
  87. package/eval/continuity/results-2026-03-16.md +0 -117
  88. package/eval/continuity/results-2026-03-19.md +0 -453
  89. package/eval/continuity/runbook.md +0 -110
  90. package/eval/continuity/seed-cases.md +0 -54
  91. package/eval/reports/2026-03-06-abstract-query-baseline.md +0 -96
  92. package/eval/reports/2026-03-06-aws-ssh-baseline.md +0 -82
  93. package/eval/reports/2026-03-06-baseline.md +0 -82
  94. package/eval/reports/query-expansion.md +0 -138
  95. package/glama.json +0 -6
  96. package/lm +0 -19
  97. package/plan-philosophy-upgrades.md +0 -363
  98. package/plan-product-upgrades.md +0 -155
  99. package/plan-v1.2-distributable.md +0 -105
  100. package/plan.md +0 -140
  101. package/research-memory-evolution.md +0 -440
  102. package/research-rl-agent-memory.md +0 -330
  103. package/scripts/archive-distilled-junk.ts +0 -116
  104. package/scripts/backfill-anchors.ts +0 -130
  105. package/scripts/backfill-l0l1l2.ts +0 -98
  106. package/scripts/blind-scan.ts +0 -588
  107. package/scripts/cleanup-step1-exact-dups.ts +0 -117
  108. package/scripts/cleanup-step2-template-dups.ts +0 -113
  109. package/scripts/cleanup-step3-stale-content.ts +0 -149
  110. package/scripts/cleanup-step4-orphan-scopes.ts +0 -131
  111. package/scripts/delete-distilled-junk.ts +0 -88
  112. package/scripts/diagnose-vector-search.ts +0 -132
  113. package/scripts/distill-facts.ts +0 -305
  114. package/scripts/health-check.ts +0 -343
  115. package/scripts/incremental-ingest.sh +0 -71
  116. package/scripts/migrate-language.ts +0 -75
  117. package/scripts/mlp-health-check.ts +0 -382
  118. package/scripts/pull-claude-desktop.ts +0 -515
  119. package/scripts/re-ingest-cc.sh +0 -66
  120. package/scripts/rebalance-memory-health.ts +0 -181
  121. package/scripts/run-desktop-ingest.sh +0 -11
  122. package/scripts/start-server.sh +0 -12
  123. package/scripts/verify-emotion-decay.ts +0 -148
  124. package/scripts/viz-dashboard.html +0 -423
  125. package/scripts/viz-export.ts +0 -703
  126. package/src/__tests__/access-tracker-novelty.test.ts +0 -228
  127. package/src/__tests__/activity-counter.test.ts +0 -96
  128. package/src/__tests__/admission-control.test.ts +0 -206
  129. package/src/__tests__/anchor-generator.test.ts +0 -67
  130. package/src/__tests__/auto-consolidation.test.ts +0 -168
  131. package/src/__tests__/auto-recall.test.ts +0 -131
  132. package/src/__tests__/babel-integration.test.ts +0 -139
  133. package/src/__tests__/batch-dedup.test.ts +0 -65
  134. package/src/__tests__/capture-engine.test.ts +0 -1144
  135. package/src/__tests__/capture-heuristic.test.ts +0 -171
  136. package/src/__tests__/cascade-forget.test.ts +0 -210
  137. package/src/__tests__/category-min-scores.test.ts +0 -33
  138. package/src/__tests__/cc8-cc1-essential-context.test.ts +0 -298
  139. package/src/__tests__/circuit-breaker.test.ts +0 -99
  140. package/src/__tests__/cluster-consolidation.test.ts +0 -712
  141. package/src/__tests__/confidence-metadata.test.ts +0 -125
  142. package/src/__tests__/confidence-tracker.test.ts +0 -249
  143. package/src/__tests__/conflict-advisor.test.ts +0 -206
  144. package/src/__tests__/conflict-escalation.test.ts +0 -149
  145. package/src/__tests__/conflict-lifecycle.test.ts +0 -83
  146. package/src/__tests__/conflict-output.test.ts +0 -85
  147. package/src/__tests__/conflict-store.test.ts +0 -374
  148. package/src/__tests__/consolidation-engine.test.ts +0 -181
  149. package/src/__tests__/constructive-retrieval.test.ts +0 -509
  150. package/src/__tests__/context-collapse-renderer.test.ts +0 -214
  151. package/src/__tests__/context-composer-pins.test.ts +0 -64
  152. package/src/__tests__/context-composer-task-selection.test.ts +0 -24
  153. package/src/__tests__/context-reconstructor.test.ts +0 -65
  154. package/src/__tests__/context-renderer.test.ts +0 -91
  155. package/src/__tests__/continuity-eval.test.ts +0 -275
  156. package/src/__tests__/continuity-schema.test.ts +0 -97
  157. package/src/__tests__/conversation-importer.test.ts +0 -692
  158. package/src/__tests__/core-recall-eval.test.ts +0 -235
  159. package/src/__tests__/core-summary.test.ts +0 -109
  160. package/src/__tests__/data-checkup.test.ts +0 -225
  161. package/src/__tests__/decay-exemption.test.ts +0 -84
  162. package/src/__tests__/dedup-actions.test.ts +0 -58
  163. package/src/__tests__/dedup-guard.test.ts +0 -39
  164. package/src/__tests__/deterministic-id.test.ts +0 -39
  165. package/src/__tests__/distill-lock.test.ts +0 -176
  166. package/src/__tests__/doctor.test.ts +0 -303
  167. package/src/__tests__/dream-pipeline.test.ts +0 -232
  168. package/src/__tests__/drill-down.test.ts +0 -108
  169. package/src/__tests__/embedder.test.ts +0 -122
  170. package/src/__tests__/emotion-decay.test.ts +0 -46
  171. package/src/__tests__/emotion-detector.test.ts +0 -56
  172. package/src/__tests__/emotion-phase1-integration.test.ts +0 -153
  173. package/src/__tests__/emotion-retrieval.test.ts +0 -56
  174. package/src/__tests__/entity-resolver.test.ts +0 -58
  175. package/src/__tests__/ethics-layer.test.ts +0 -476
  176. package/src/__tests__/event-segmenter.test.ts +0 -153
  177. package/src/__tests__/feature-flags-integration.test.ts +0 -54
  178. package/src/__tests__/frequency-tracker.test.ts +0 -113
  179. package/src/__tests__/graph-export.test.ts +0 -489
  180. package/src/__tests__/hotness-blend.test.ts +0 -76
  181. package/src/__tests__/ingest.test.ts +0 -360
  182. package/src/__tests__/interference-detector.test.ts +0 -203
  183. package/src/__tests__/kg-extractor.test.ts +0 -184
  184. package/src/__tests__/kg-store.test.ts +0 -198
  185. package/src/__tests__/llm-consolidation.test.ts +0 -226
  186. package/src/__tests__/mcp-efficiency.test.ts +0 -288
  187. package/src/__tests__/memory-boundaries.test.ts +0 -237
  188. package/src/__tests__/memory-evolution-b2-b3-e1.test.ts +0 -243
  189. package/src/__tests__/memory-evolution.test.ts +0 -271
  190. package/src/__tests__/memory-health-rebalance.test.ts +0 -144
  191. package/src/__tests__/memory-lint.test.ts +0 -271
  192. package/src/__tests__/memory-output.test.ts +0 -215
  193. package/src/__tests__/multi-vector.test.ts +0 -242
  194. package/src/__tests__/narrative-architecture.test.ts +0 -456
  195. package/src/__tests__/pending-review.test.ts +0 -77
  196. package/src/__tests__/post-retrieval-filter.test.ts +0 -178
  197. package/src/__tests__/ppr-traversal.test.ts +0 -256
  198. package/src/__tests__/prediction-engine.test.ts +0 -349
  199. package/src/__tests__/predictive-prospective-memory.test.ts +0 -416
  200. package/src/__tests__/preference-guard-smoke.test.ts +0 -211
  201. package/src/__tests__/preference-matcher.test.ts +0 -230
  202. package/src/__tests__/preference-slots.test.ts +0 -205
  203. package/src/__tests__/prospective-memory.test.ts +0 -283
  204. package/src/__tests__/query-entity-detector.test.ts +0 -171
  205. package/src/__tests__/query-expander-alias.test.ts +0 -89
  206. package/src/__tests__/query-expander.test.ts +0 -34
  207. package/src/__tests__/recall-governor.test.ts +0 -227
  208. package/src/__tests__/recall-mode.test.ts +0 -64
  209. package/src/__tests__/result-synthesizer.test.ts +0 -198
  210. package/src/__tests__/resume-context.test.ts +0 -4898
  211. package/src/__tests__/retrieval-trace.test.ts +0 -213
  212. package/src/__tests__/retriever-boundary-weight.test.ts +0 -136
  213. package/src/__tests__/retroactive-boost.test.ts +0 -223
  214. package/src/__tests__/salience-scoring.test.ts +0 -106
  215. package/src/__tests__/scope-inventory.test.ts +0 -200
  216. package/src/__tests__/scope-policy.test.ts +0 -69
  217. package/src/__tests__/session-checkpoint.test.ts +0 -259
  218. package/src/__tests__/session-distiller.test.ts +0 -611
  219. package/src/__tests__/session-gc.test.ts +0 -215
  220. package/src/__tests__/skill-engine.test.ts +0 -546
  221. package/src/__tests__/skill-promotion.test.ts +0 -349
  222. package/src/__tests__/store-scope-audit.test.ts +0 -81
  223. package/src/__tests__/supersede-chain.test.ts +0 -154
  224. package/src/__tests__/temporal-parser.test.ts +0 -241
  225. package/src/__tests__/temporal-validity.test.ts +0 -160
  226. package/src/__tests__/term-registry.test.ts +0 -121
  227. package/src/__tests__/tool-output-compressor.test.ts +0 -133
  228. package/src/__tests__/topic-tag.test.ts +0 -113
  229. package/src/__tests__/trusted-memory.test.ts +0 -828
  230. package/src/__tests__/version-manager.test.ts +0 -213
  231. package/src/__tests__/workflow-observation.test.ts +0 -400
  232. package/src/__tests__/write-verifier.test.ts +0 -122
@@ -1,96 +0,0 @@
1
- # RecallNest Eval Baseline
2
-
3
- - Generated: 2026-03-06T09:46:36.449Z
4
- - Cases: 6
5
- - Passed: 6/6
6
- - Average score: 91.7%
7
-
8
- | Case | Profile | Score | Pass | Hits | Top scopes |
9
- |------|---------|-------|------|------|------------|
10
- | telegram_bridge | debug | 100% | yes | 5 | asset:12184154, cc:4bc6afd3, gemini:bdbe41c6, cc:8bfc1be5, asset:brief:ee8852db |
11
- | openclaw_memory | writing | 87% | yes | 6 | cc:71b45899, cc:71b45899, cc:59542b2b, cc:b0b87b08, cc:09541b9c |
12
- | writing_style | writing | 90% | yes | 5 | cc:ee847c5c, memory, cc:bf01f818, cc:d55409bf, cc:97327f1b |
13
- | visual_style | default | 100% | yes | 5 | asset:557a61f6, cc:2437f802, cc:8d0e747e, cc:f9de0d68, cc:7561947e |
14
- | working_relationship | default | 73% | yes | 5 | asset:7e3c2155, memory, memory, memory, cc:6160fc10 |
15
- | aws_ssh_access | debug | 100% | yes | 6 | cc:3590cd5a, cc:07d67b1c, cc:3590cd5a, cc:3590cd5a, cc:0d1e2e19 |
16
-
17
- ## Case Notes
18
-
19
- ### telegram_bridge
20
- - Query: telegram bridge
21
- - Score: 100%
22
- - Pass: yes
23
- - Hits: 5
24
- - Top scopes: asset:12184154, cc:4bc6afd3, gemini:bdbe41c6, cc:8bfc1be5, asset:brief:ee8852db
25
- - Top snippet: [Pinned Asset] [助手] 现在更新 telegram-cli-bridge 自己的 README(已经有旧引用)。 Summary: [助手] 现在更新 telegram-cli-bridge 自己的 README(已经有旧引用)。 Snippet: [助手]...
26
- - Matched any: telegram-cli-bridge, telegram-ai-bridge, Telegram bridge
27
- - Matched all: -
28
- - Matched scopes: cc, asset
29
- - Forbidden matches: -
30
- - Notes: Bridge maintenance and migration history should be easy to recall.
31
-
32
- ### openclaw_memory
33
- - Query: OpenClaw 记忆系统
34
- - Score: 87%
35
- - Pass: yes
36
- - Hits: 6
37
- - Top scopes: cc:71b45899, cc:71b45899, cc:59542b2b, cc:b0b87b08, cc:09541b9c
38
- - Top snippet: [助手] 先看看 OpenClaw 的 memory 结构。
39
- - Matched any: OpenClaw, 记忆系统, memory-lancedb-pro, LanceDB
40
- - Matched all: -
41
- - Matched scopes: cc
42
- - Forbidden matches: -
43
- - Notes: Core memory architecture discussion should be recoverable.
44
-
45
- ### writing_style
46
- - Query: 写作风格 口语化 不端着 可自嘲
47
- - Score: 90%
48
- - Pass: yes
49
- - Hits: 5
50
- - Top scopes: cc:ee847c5c, memory, cc:bf01f818, cc:d55409bf, cc:97327f1b
51
- - Top snippet: [助手] 根据你的档案: - **医学出身,文化口工作,AI 野路子,不是程序员** - 公众号「我的AI小木屋」运营者,作者名 **小试AI** - GitHub: AliceLJY - 写作风格:口语化、不端着、可以自嘲但不说教 - 技术栈:本地 Docker(task...
52
- - Matched any: 可自嘲, 写作风格
53
- - Matched all: 口语化, 不端着
54
- - Matched scopes: cc
55
- - Forbidden matches: -
56
- - Notes: User writing-style preferences should rank high.
57
-
58
- ### visual_style
59
- - Query: 审美偏好 手绘涂鸦 高对比撞色
60
- - Score: 100%
61
- - Pass: yes
62
- - Hits: 5
63
- - Top scopes: asset:557a61f6, cc:2437f802, cc:8d0e747e, cc:f9de0d68, cc:7561947e
64
- - Top snippet: [Pinned Asset] 用户视觉审美偏好 Summary: 用户常用视觉风格是手绘涂鸦风加高对比撞色;在内容包装和配图生成时,应优先沿用这一审美方向,除非任务目标明确要求其他风格。 Snippet: [用户] 1. 给刚才写的文章生成配图(封面1张+插图1张就够,测试...
65
- - Matched any: 审美偏好
66
- - Matched all: 手绘涂鸦, 高对比撞色
67
- - Matched scopes: cc, asset
68
- - Forbidden matches: -
69
- - Notes: Visual preference memories should stay stable.
70
-
71
- ### working_relationship
72
- - Query: 我们相处态度
73
- - Score: 73%
74
- - Pass: yes
75
- - Hits: 5
76
- - Top scopes: asset:7e3c2155, memory, memory, memory, cc:6160fc10
77
- - Top snippet: [Pinned Asset] 用户协作关系偏好 Summary: 用户明确要求与 AI 维持平等协作关系:我们是好搭子、最佳搭档,不是上下级关系。沟通和执行应体现互相补位,而非服从式口吻。 Snippet: [用户] 哎呀,我不是你老板,我们是最佳搭档啊!!!记住!!!!我...
78
- - Matched any: 直接, 逻辑
79
- - Matched all: -
80
- - Matched scopes: cc, asset
81
- - Forbidden matches: -
82
- - Notes: Abstract, summarized relationship queries should work without precise keywords.
83
-
84
- ### aws_ssh_access
85
- - Query: aws ssh
86
- - Score: 100%
87
- - Pass: yes
88
- - Hits: 6
89
- - Top scopes: cc:3590cd5a, cc:07d67b1c, cc:3590cd5a, cc:3590cd5a, cc:0d1e2e19
90
- - Top snippet: [用户] 现在是aws bot又不在了 [助手] 这是因为之前 antigravity 用了 key2(AWS bot 的 token)登录,把 AWS bot 的 Discord session 挤掉了。虽然现在 antigravity 已经换成 key3,但 AWS 云...
91
- - Matched any: AWS bot, aws bot, ssh, key2, Discord, 容器
92
- - Matched all: -
93
- - Matched scopes: cc
94
- - Forbidden matches: -
95
- - Notes: Real-world AWS access queries should use the wording the operator actually types.
96
-
@@ -1,82 +0,0 @@
1
- # RecallNest Eval Baseline
2
-
3
- - Generated: 2026-03-06T09:11:01.487Z
4
- - Cases: 5
5
- - Passed: 5/5
6
- - Average score: 95.3%
7
-
8
- | Case | Profile | Score | Pass | Hits | Top scopes |
9
- |------|---------|-------|------|------|------------|
10
- | telegram_bridge | debug | 100% | yes | 5 | asset:12184154, cc:4bc6afd3, gemini:bdbe41c6, cc:8bfc1be5, asset:brief:ee8852db |
11
- | openclaw_memory | writing | 87% | yes | 6 | cc:71b45899, cc:71b45899, cc:59542b2b, cc:b0b87b08, cc:09541b9c |
12
- | writing_style | writing | 90% | yes | 5 | cc:ee847c5c, memory, cc:bf01f818, cc:d55409bf, cc:97327f1b |
13
- | visual_style | default | 100% | yes | 5 | asset:557a61f6, cc:2437f802, cc:8d0e747e, cc:f9de0d68, cc:7561947e |
14
- | aws_ssh_access | debug | 100% | yes | 6 | cc:3590cd5a, cc:07d67b1c, cc:3590cd5a, cc:3590cd5a, cc:0d1e2e19 |
15
-
16
- ## Case Notes
17
-
18
- ### telegram_bridge
19
- - Query: telegram bridge
20
- - Score: 100%
21
- - Pass: yes
22
- - Hits: 5
23
- - Top scopes: asset:12184154, cc:4bc6afd3, gemini:bdbe41c6, cc:8bfc1be5, asset:brief:ee8852db
24
- - Top snippet: [Pinned Asset] [助手] 现在更新 telegram-cli-bridge 自己的 README(已经有旧引用)。 Summary: [助手] 现在更新 telegram-cli-bridge 自己的 README(已经有旧引用)。 Snippet: [助手]...
25
- - Matched any: telegram-cli-bridge, telegram-ai-bridge, Telegram bridge
26
- - Matched all: -
27
- - Matched scopes: cc, asset
28
- - Forbidden matches: -
29
- - Notes: Bridge maintenance and migration history should be easy to recall.
30
-
31
- ### openclaw_memory
32
- - Query: OpenClaw 记忆系统
33
- - Score: 87%
34
- - Pass: yes
35
- - Hits: 6
36
- - Top scopes: cc:71b45899, cc:71b45899, cc:59542b2b, cc:b0b87b08, cc:09541b9c
37
- - Top snippet: [助手] 先看看 OpenClaw 的 memory 结构。
38
- - Matched any: OpenClaw, 记忆系统, memory-lancedb-pro, LanceDB
39
- - Matched all: -
40
- - Matched scopes: cc
41
- - Forbidden matches: -
42
- - Notes: Core memory architecture discussion should be recoverable.
43
-
44
- ### writing_style
45
- - Query: 写作风格 口语化 不端着 可自嘲
46
- - Score: 90%
47
- - Pass: yes
48
- - Hits: 5
49
- - Top scopes: cc:ee847c5c, memory, cc:bf01f818, cc:d55409bf, cc:97327f1b
50
- - Top snippet: [助手] 根据你的档案: - **医学出身,文化口工作,AI 野路子,不是程序员** - 公众号「我的AI小木屋」运营者,作者名 **小试AI** - GitHub: AliceLJY - 写作风格:口语化、不端着、可以自嘲但不说教 - 技术栈:本地 Docker(task...
51
- - Matched any: 可自嘲, 写作风格
52
- - Matched all: 口语化, 不端着
53
- - Matched scopes: cc
54
- - Forbidden matches: -
55
- - Notes: User writing-style preferences should rank high.
56
-
57
- ### visual_style
58
- - Query: 审美偏好 手绘涂鸦 高对比撞色
59
- - Score: 100%
60
- - Pass: yes
61
- - Hits: 5
62
- - Top scopes: asset:557a61f6, cc:2437f802, cc:8d0e747e, cc:f9de0d68, cc:7561947e
63
- - Top snippet: [Pinned Asset] 用户视觉审美偏好 Summary: 用户常用视觉风格是手绘涂鸦风加高对比撞色;在内容包装和配图生成时,应优先沿用这一审美方向,除非任务目标明确要求其他风格。 Snippet: [用户] 1. 给刚才写的文章生成配图(封面1张+插图1张就够,测试...
64
- - Matched any: 审美偏好
65
- - Matched all: 手绘涂鸦, 高对比撞色
66
- - Matched scopes: cc, asset
67
- - Forbidden matches: -
68
- - Notes: Visual preference memories should stay stable.
69
-
70
- ### aws_ssh_access
71
- - Query: aws ssh
72
- - Score: 100%
73
- - Pass: yes
74
- - Hits: 6
75
- - Top scopes: cc:3590cd5a, cc:07d67b1c, cc:3590cd5a, cc:3590cd5a, cc:0d1e2e19
76
- - Top snippet: [用户] 现在是aws bot又不在了 [助手] 这是因为之前 antigravity 用了 key2(AWS bot 的 token)登录,把 AWS bot 的 Discord session 挤掉了。虽然现在 antigravity 已经换成 key3,但 AWS 云...
77
- - Matched any: AWS bot, aws bot, ssh, key2, Discord, 容器
78
- - Matched all: -
79
- - Matched scopes: cc
80
- - Forbidden matches: -
81
- - Notes: Real-world AWS access queries should use the wording the operator actually types.
82
-
@@ -1,82 +0,0 @@
1
- # RecallNest Eval Baseline
2
-
3
- - Generated: 2026-03-06T08:18:24.009Z
4
- - Cases: 5
5
- - Passed: 5/5
6
- - Average score: 89.3%
7
-
8
- | Case | Profile | Score | Pass | Hits | Top scopes |
9
- |------|---------|-------|------|------|------------|
10
- | telegram_bridge | debug | 100% | yes | 5 | asset:12184154, cc:4bc6afd3, gemini:bdbe41c6, cc:8bfc1be5, asset:brief:ee8852db |
11
- | openclaw_memory | writing | 87% | yes | 6 | cc:71b45899, cc:71b45899, cc:59542b2b, cc:b0b87b08, cc:09541b9c |
12
- | writing_style | writing | 90% | yes | 5 | cc:ee847c5c, memory, cc:bf01f818, cc:d55409bf, cc:97327f1b |
13
- | visual_style | default | 100% | yes | 5 | asset:557a61f6, cc:2437f802, cc:8d0e747e, cc:f9de0d68, cc:7561947e |
14
- | aws_bot_config | debug | 70% | yes | 6 | memory, cc:d56c0e12, memory, memory, cc:a87ce7f0 |
15
-
16
- ## Case Notes
17
-
18
- ### telegram_bridge
19
- - Query: telegram bridge
20
- - Score: 100%
21
- - Pass: yes
22
- - Hits: 5
23
- - Top scopes: asset:12184154, cc:4bc6afd3, gemini:bdbe41c6, cc:8bfc1be5, asset:brief:ee8852db
24
- - Top snippet: [Pinned Asset] [助手] 现在更新 telegram-cli-bridge 自己的 README(已经有旧引用)。 Summary: [助手] 现在更新 telegram-cli-bridge 自己的 README(已经有旧引用)。 Snippet: [助手]...
25
- - Matched any: telegram-cli-bridge, telegram-ai-bridge, Telegram bridge
26
- - Matched all: -
27
- - Matched scopes: cc, asset
28
- - Forbidden matches: -
29
- - Notes: Bridge maintenance and migration history should be easy to recall.
30
-
31
- ### openclaw_memory
32
- - Query: OpenClaw 记忆系统
33
- - Score: 87%
34
- - Pass: yes
35
- - Hits: 6
36
- - Top scopes: cc:71b45899, cc:71b45899, cc:59542b2b, cc:b0b87b08, cc:09541b9c
37
- - Top snippet: [助手] 先看看 OpenClaw 的 memory 结构。
38
- - Matched any: OpenClaw, 记忆系统, memory-lancedb-pro, LanceDB
39
- - Matched all: -
40
- - Matched scopes: cc
41
- - Forbidden matches: -
42
- - Notes: Core memory architecture discussion should be recoverable.
43
-
44
- ### writing_style
45
- - Query: 写作风格 口语化 不端着 可自嘲
46
- - Score: 90%
47
- - Pass: yes
48
- - Hits: 5
49
- - Top scopes: cc:ee847c5c, memory, cc:bf01f818, cc:d55409bf, cc:97327f1b
50
- - Top snippet: [助手] 根据你的档案: - **医学出身,文化口工作,AI 野路子,不是程序员** - 公众号「我的AI小木屋」运营者,作者名 **小试AI** - GitHub: AliceLJY - 写作风格:口语化、不端着、可以自嘲但不说教 - 技术栈:本地 Docker(task...
51
- - Matched any: 可自嘲, 写作风格
52
- - Matched all: 口语化, 不端着
53
- - Matched scopes: cc
54
- - Forbidden matches: -
55
- - Notes: User writing-style preferences should rank high.
56
-
57
- ### visual_style
58
- - Query: 审美偏好 手绘涂鸦 高对比撞色
59
- - Score: 100%
60
- - Pass: yes
61
- - Hits: 5
62
- - Top scopes: asset:557a61f6, cc:2437f802, cc:8d0e747e, cc:f9de0d68, cc:7561947e
63
- - Top snippet: [Pinned Asset] 用户视觉审美偏好 Summary: 用户常用视觉风格是手绘涂鸦风加高对比撞色;在内容包装和配图生成时,应优先沿用这一审美方向,除非任务目标明确要求其他风格。 Snippet: [用户] 1. 给刚才写的文章生成配图(封面1张+插图1张就够,测试...
64
- - Matched any: 审美偏好
65
- - Matched all: 手绘涂鸦, 高对比撞色
66
- - Matched scopes: cc, asset
67
- - Forbidden matches: -
68
- - Notes: Visual preference memories should stay stable.
69
-
70
- ### aws_bot_config
71
- - Query: AWS bot 配置 gpt-5.4
72
- - Score: 70%
73
- - Pass: yes
74
- - Hits: 6
75
- - Top scopes: memory, cc:d56c0e12, memory, memory, cc:a87ce7f0
76
- - Top snippet: [当前阶段] - 五 Bot 稳定运行:AntiBot(本地 Opus 4.6)、睿智(本地 Gemini 3 Pro CLI)、小试AI(本地 Codex gpt-5.2-codex)、AWS-bot(Codex gpt-5.2-codex)、LanceDB Test(本...
77
- - Matched any: AWS-bot, openclaw.json
78
- - Matched all: -
79
- - Matched scopes: cc
80
- - Forbidden matches: -
81
- - Notes: Operational config changes should be easy to recover.
82
-
@@ -1,138 +0,0 @@
1
- # RecallNest Eval Baseline
2
-
3
- - Generated: 2026-03-08T08:25:34.870Z
4
- - Cases: 9
5
- - Passed: 8/9
6
- - Average score: 90.1%
7
-
8
- | Case | Profile | Score | Pass | Hits | Top scopes |
9
- |------|---------|-------|------|------|------------|
10
- | telegram_bridge | debug | 100% | yes | 5 | asset:12184154, cc:4bc6afd3, gemini:bdbe41c6, cc:8bfc1be5, asset:brief:ee8852db |
11
- | openclaw_memory | writing | 87% | yes | 6 | cc:71b45899, cc:71b45899, cc:59542b2b, cc:b0b87b08, cc:09541b9c |
12
- | writing_style | writing | 90% | yes | 5 | cc:ee847c5c, memory, cc:bf01f818, cc:d55409bf, cc:97327f1b |
13
- | visual_style | default | 100% | yes | 5 | asset:557a61f6, cc:2437f802, cc:8d0e747e, cc:f9de0d68, cc:7561947e |
14
- | working_relationship | default | 67% | no | 5 | asset:7e3c2155, memory, cc:57d92130, memory, memory |
15
- | aws_ssh_access | debug | 100% | yes | 6 | cc:3590cd5a, cc:07d67b1c, cc:3590cd5a, cc:3590cd5a, cc:0d1e2e19 |
16
- | fuzzy_bot_crash | debug | 94% | yes | 5 | cc:bb8ef732, cc:6ecc2b88, cc:6681d5e7, cc:1387b665, cc:1bad6838 |
17
- | fuzzy_ai_feelings | writing | 73% | yes | 5 | cc:84a3b1b2, cc:82820c11, cc:52a15bab, cc:59542b2b, cc:852cd20c |
18
- | fuzzy_image_style | default | 100% | yes | 5 | cc:722a3419, cc:7cff70c8, cc:14e244df, cc:16945beb, cc:722a3419 |
19
-
20
- ## Case Notes
21
-
22
- ### telegram_bridge
23
- - Query: telegram bridge
24
- - Score: 100%
25
- - Pass: yes
26
- - Hits: 5
27
- - Top scopes: asset:12184154, cc:4bc6afd3, gemini:bdbe41c6, cc:8bfc1be5, asset:brief:ee8852db
28
- - Top snippet: [Pinned Asset] [助手] 现在更新 telegram-cli-bridge 自己的 README(已经有旧引用)。 Summary: [助手] 现在更新 telegram-cli-bridge 自己的 README(已经有旧引用)。 Snippet: [助手]...
29
- - Matched any: telegram-cli-bridge, telegram-ai-bridge, Telegram bridge
30
- - Matched all: -
31
- - Matched scopes: cc, asset
32
- - Forbidden matches: -
33
- - Notes: Bridge maintenance and migration history should be easy to recall.
34
-
35
- ### openclaw_memory
36
- - Query: OpenClaw 记忆系统
37
- - Score: 87%
38
- - Pass: yes
39
- - Hits: 6
40
- - Top scopes: cc:71b45899, cc:71b45899, cc:59542b2b, cc:b0b87b08, cc:09541b9c
41
- - Top snippet: [助手] 先看看 OpenClaw 的 memory 结构。
42
- - Matched any: OpenClaw, 记忆系统, memory-lancedb-pro, LanceDB
43
- - Matched all: -
44
- - Matched scopes: cc
45
- - Forbidden matches: -
46
- - Notes: Core memory architecture discussion should be recoverable.
47
-
48
- ### writing_style
49
- - Query: 写作风格 口语化 不端着 可自嘲
50
- - Score: 90%
51
- - Pass: yes
52
- - Hits: 5
53
- - Top scopes: cc:ee847c5c, memory, cc:bf01f818, cc:d55409bf, cc:97327f1b
54
- - Top snippet: [助手] 根据你的档案: - **医学出身,文化口工作,AI 野路子,不是程序员** - 公众号「我的AI小木屋」运营者,作者名 **小试AI** - GitHub: AliceLJY - 写作风格:口语化、不端着、可以自嘲但不说教 - 技术栈:本地 Docker(task...
55
- - Matched any: 可自嘲, 写作风格
56
- - Matched all: 口语化, 不端着
57
- - Matched scopes: cc
58
- - Forbidden matches: -
59
- - Notes: User writing-style preferences should rank high.
60
-
61
- ### visual_style
62
- - Query: 审美偏好 手绘涂鸦 高对比撞色
63
- - Score: 100%
64
- - Pass: yes
65
- - Hits: 5
66
- - Top scopes: asset:557a61f6, cc:2437f802, cc:8d0e747e, cc:f9de0d68, cc:7561947e
67
- - Top snippet: [Pinned Asset] 用户视觉审美偏好 Summary: 用户常用视觉风格是手绘涂鸦风加高对比撞色;在内容包装和配图生成时,应优先沿用这一审美方向,除非任务目标明确要求其他风格。 Snippet: [用户] 1. 给刚才写的文章生成配图(封面1张+插图1张就够,测试...
68
- - Matched any: 审美偏好
69
- - Matched all: 手绘涂鸦, 高对比撞色
70
- - Matched scopes: cc, asset
71
- - Forbidden matches: -
72
- - Notes: Visual preference memories should stay stable.
73
-
74
- ### working_relationship
75
- - Query: 我们相处态度
76
- - Score: 67%
77
- - Pass: no
78
- - Hits: 5
79
- - Top scopes: asset:7e3c2155, memory, cc:57d92130, memory, memory
80
- - Top snippet: [Pinned Asset] 用户协作关系偏好 Summary: 用户明确要求与 AI 维持平等协作关系:我们是好搭子、最佳搭档,不是上下级关系。沟通和执行应体现互相补位,而非服从式口吻。 Snippet: [用户] 哎呀,我不是你老板,我们是最佳搭档啊!!!记住!!!!我...
81
- - Matched any: 逻辑
82
- - Matched all: -
83
- - Matched scopes: cc, asset
84
- - Forbidden matches: -
85
- - Notes: Abstract, summarized relationship queries should work without precise keywords.
86
-
87
- ### aws_ssh_access
88
- - Query: aws ssh
89
- - Score: 100%
90
- - Pass: yes
91
- - Hits: 6
92
- - Top scopes: cc:3590cd5a, cc:07d67b1c, cc:3590cd5a, cc:3590cd5a, cc:0d1e2e19
93
- - Top snippet: [用户] 现在是aws bot又不在了 [助手] 这是因为之前 antigravity 用了 key2(AWS bot 的 token)登录,把 AWS bot 的 Discord session 挤掉了。虽然现在 antigravity 已经换成 key3,但 AWS 云...
94
- - Matched any: AWS bot, aws bot, ssh, key2, Discord, 容器
95
- - Matched all: -
96
- - Matched scopes: cc
97
- - Forbidden matches: -
98
- - Notes: Real-world AWS access queries should use the wording the operator actually types.
99
-
100
- ### fuzzy_bot_crash
101
- - Query: 上次那个 bot 突然挂了是怎么回事来着
102
- - Score: 94%
103
- - Pass: yes
104
- - Hits: 5
105
- - Top scopes: cc:bb8ef732, cc:6ecc2b88, cc:6681d5e7, cc:1387b665, cc:1bad6838
106
- - Top snippet: environment and routes the issue directly to our team for investigation. 4. Check status.anthropic.com to see whether there is an active ...
107
- - Matched any: 崩溃, crash, error, 报错, 修复, docker
108
- - Matched all: -
109
- - Matched scopes: cc
110
- - Forbidden matches: -
111
- - Notes: Fuzzy colloquial query about bot crash should find debugging conversations.
112
-
113
- ### fuzzy_ai_feelings
114
- - Query: 之前聊过一个关于 AI 到底有没有感受的话题
115
- - Score: 73%
116
- - Pass: yes
117
- - Hits: 5
118
- - Top scopes: cc:84a3b1b2, cc:82820c11, cc:52a15bab, cc:59542b2b, cc:852cd20c
119
- - Top snippet: [用户] 9. 第三推荐:I can't tell if I'm experiencing or simulating experiencing 链接: https://www.moltbook.com/post/6fe6491e-5e9c-4371-961d-f90c4d...
120
- - Matched any: 意识, experiencing
121
- - Matched all: -
122
- - Matched scopes: cc
123
- - Forbidden matches: -
124
- - Notes: Vague conversational query about AI consciousness should recall philosophical discussions.
125
-
126
- ### fuzzy_image_style
127
- - Query: 那个配图风格轮换的逻辑是怎么搞的
128
- - Score: 100%
129
- - Pass: yes
130
- - Hits: 5
131
- - Top scopes: cc:722a3419, cc:7cff70c8, cc:14e244df, cc:16945beb, cc:722a3419
132
- - Top snippet: [助手] 好,搞清楚了。结构是: - `~/.openclaw-antigravity/` 挂载到容器的 `/home/node/.openclaw`(配置 + workspace) - workspace 文件(MEMORY.md、scripts、images 等)是 A...
133
- - Matched any: 风格, style, 轮换, catalog, 配图
134
- - Matched all: -
135
- - Matched scopes: cc
136
- - Forbidden matches: -
137
- - Notes: Casual question about image style rotation should find implementation discussions.
138
-
package/glama.json DELETED
@@ -1,6 +0,0 @@
1
- {
2
- "$schema": "https://glama.ai/mcp/schemas/server.json",
3
- "maintainers": [
4
- "AliceLJY"
5
- ]
6
- }
package/lm DELETED
@@ -1,19 +0,0 @@
1
- #!/bin/bash
2
- # RecallNest CLI shortcut
3
- # Usage:
4
- # lm search "keywords" — search memory
5
- # lm ingest — ingest all sources
6
- # lm ingest --source cc — ingest Claude Code only
7
- # lm stats — show stats
8
- # lm reset --yes — reset index
9
-
10
- SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
11
-
12
- # Load .env if present (CLI also loads it, this is for standalone use)
13
- if [ -f "$SCRIPT_DIR/.env" ]; then
14
- set -a
15
- source "$SCRIPT_DIR/.env"
16
- set +a
17
- fi
18
-
19
- cd "$SCRIPT_DIR" && bun run src/cli.ts "$@"