feitianxiang
ad84e5decd
Merge pull request #344 from Tencent-Hunyuan/fix/memhub-hooks-no-evidence
...
fix(memhub): capture hooks hand turns to Core instead of writing Evidence
2026-11-05 16:03:24 +08:00
feitianxiang
b0e429b9ae
Merge pull request #322 from Tencent-Hunyuan/docs/memhub-1.0.6-changelog
...
docs(changelog): complete MemHub 1.0.6 notes (build 41)
2026-11-04 20:49:02 +08:00
feitianxiang
837c084be2
Merge pull request #320 from Tencent-Hunyuan/fix/memhub-db-contention
...
fix(memhub): bound import jobs, keep the write queue off the bulk pool, nothing blocking on the event loop
2026-11-04 20:27:32 +08:00
feitianxiang
964d8610f3
Merge pull request #319 from Tencent-Hunyuan/fix/memhub-lite-raw-rag
...
fix(memhub): hy-memory 2.0.14; Lite search is plain RAG over raw
2026-11-04 19:42:51 +08:00
feitianxiang
e1933d3307
Merge pull request #318 from Tencent-Hunyuan/release/hy-memory-2.0.14
...
release: hy-memory 2.0.14
2026-11-04 19:42:39 +08:00
feitianxiang
d2e9cdaf06
Merge pull request #286 from Tencent-Hunyuan/chore/memhub-build-38
...
chore(release): MemHub 1.0.6, build 38
2026-11-02 21:47:36 +08:00
feitianxiang
65550621e0
Merge pull request #285 from Tencent-Hunyuan/release/memhub-1.0.6-hy-2.0.13
...
chore(memhub): hy-memory 2.0.13; 1.0.6 changelog
2026-11-02 21:29:36 +08:00
feitianxiang
58f1477680
Merge pull request #282 from Tencent-Hunyuan/release/hy-memory-2.0.13
...
release: hy-memory 2.0.13
2026-11-02 21:20:29 +08:00
feitianxiang
2483412475
Merge pull request #281 from Tencent-Hunyuan/fix/sdk-write-failures-not-silent
...
fix(sdk): a model failure during a write reports failure and stores nothing
2026-11-02 17:29:10 +08:00
feitianxiang
f9ee983790
Merge pull request #278 from Tencent-Hunyuan/fix/queue-wal-and-recall-pools
...
fix(memhub): memory writes retry themselves, a broken model provider pauses the queue, WAL stays bounded, recall no longer starves
2026-11-02 17:28:56 +08:00
feitianxiang
850c728c9d
Merge pull request #280 from Tencent-Hunyuan/fix/multi-user-hardening
...
fix(memhub): multi-user responses echo user_id, cross-user delete answers 404, Hermes profiles isolated for any root
2026-11-02 16:44:04 +08:00
feitianxiang
34f406e96d
Merge pull request #277 from Tencent-Hunyuan/feat/memhub-hy-memory-2.0.12
...
feat(memhub): hy-memory 2.0.12; intention dedup switch in memory features
2026-11-02 15:16:36 +08:00
feitianxiang
4b83b79b2b
Merge pull request #275 from Tencent-Hunyuan/release/hy-memory-2.0.12
...
release: hy-memory 2.0.12
2026-11-02 14:58:42 +08:00
feitianxiang
3511153f9f
Merge pull request #274 from Tencent-Hunyuan/fix/deepseek-harness-profiles
...
fix(memhub): DeepSeek Harness one switch grants every profile; import versioned sessions and v3 tool text
2026-11-02 14:17:44 +08:00
feitianxiang
23facf75d5
Merge pull request #273 from Tencent-Hunyuan/fix/single-runtime-quit
...
fix(memhub): one runtime; App quit never stops the background service, stops every runtime when it is disabled
2026-11-02 14:17:32 +08:00
feitianxiang
d257188d09
Merge pull request #270 from Tencent-Hunyuan/fix/commit-validation-and-zh-language
...
fix(memhub): one commit format at the packaged-runtime boundary; zh-Hans UI sets the SDK language
2026-11-02 14:03:21 +08:00
feitianxiang
bde0220373
Merge pull request #268 from Tencent-Hunyuan/feat/sdk-intention-dedup-switch
...
feat(sdk): configurable intention dedup (on by default, unchanged behaviour)
2026-11-02 14:03:04 +08:00
feitianxiang
670c7c6cb1
Merge pull request #267 from Tencent-Hunyuan/docs/release-1.0.6
...
docs: prepare MemHub v1.0.6 (news, changelog)
2026-11-02 11:50:58 +08:00
feitianxiang
37d8662b93
Merge pull request #266 from Tencent-Hunyuan/chore/build-37
...
chore(release): MemHub 1.0.6, build 37
2026-11-02 11:44:12 +08:00
feitianxiang
2c6f269170
Merge pull request #264 from Tencent-Hunyuan/chore/pin-hy-memory-2.0.11
...
chore(memhub): pin hy-memory 2.0.11 and set the SDK language directly
2026-11-02 11:34:45 +08:00
feitianxiang
35128f2c84
Merge pull request #262 from Tencent-Hunyuan/feat/memhub-eleven-languages
...
feat(memhub): 11 display languages; non-zh/en choices set the SDK prompt language
2026-11-01 22:17:21 +08:00
feitianxiang
f12a95ce7a
Merge pull request #263 from Tencent-Hunyuan/release/hy-memory-2.0.11
...
release: hy-memory 2.0.11
2026-11-01 22:13:44 +08:00
feitianxiang
75ac1b01f2
Merge pull request #261 from Tencent-Hunyuan/feat/sdk-multilingual-prompts
...
feat(sdk): prompts in 11 languages and an explicit language option
2026-11-01 21:53:38 +08:00
feitianxiang
b2c91d9751
Merge pull request #260 from Tencent-Hunyuan/feat/memhub-embedding-switch
...
feat(memhub): switch the embedding model safely — space registry, paused re-embed, startup merge of stray spaces
2026-11-01 21:34:53 +08:00
feitianxiang
1520b6de1f
Merge pull request #259 from Tencent-Hunyuan/feat/memhub-sdk-feature-settings
...
feat(memhub): configurable SDK retrieval and extraction features (graph rerank, episode recall, …) via CLI and API
2026-11-01 20:34:32 +08:00
feitianxiang
98bc52b9f4
Merge pull request #253 from Tencent-Hunyuan/feat/memhub-mode-evolve
...
feat(memhub): memory mode (lite/pro/ultra), evolve and nightly schedule (B, stacked on #252 )
2026-11-01 19:53:15 +08:00
feitianxiang
30fea9394f
Merge pull request #252 from Tencent-Hunyuan/feat/memhub-sdk-surface
...
feat(memhub): SDK search params, memory layers, raw, v2 source aliases, profile versions (A)
2026-11-01 19:53:02 +08:00
feitianxiang
e966e913b6
Merge pull request #258 from Tencent-Hunyuan/release/hy-memory-2.0.10
...
release: hy-memory 2.0.10
2026-11-01 19:32:15 +08:00
feitianxiang
fdc57822b5
Merge pull request #254 from Tencent-Hunyuan/feat/sdk-trimmed-raw
...
feat(sdk): store raw conversations already trimmed; history concatenates without re-trimming
2026-11-01 19:24:43 +08:00
feitianxiang
104fbcce09
Merge pull request #256 from Tencent-Hunyuan/feat/sdk-reembed
...
feat(sdk): re-embed vector stores and the graph into a new embedding space
2026-11-01 18:54:40 +08:00
feitianxiang
cdb44658cf
Merge pull request #255 from Tencent-Hunyuan/fix/upgrade-migration-issues
...
fix(memhub): upgrade migration without a model profile, tolerant readback, grant triggers after cutover, reads use the active embedding
2026-11-01 18:26:00 +08:00
feitianxiang
f5ea0af75d
Merge pull request #251 from Tencent-Hunyuan/fix/hermes-injected-content
...
feat(memhub): land #250 on main — multi-user Runtime, hy-memory 2.0.9
2026-11-01 14:39:43 +08:00
feitianxiang
d29615abeb
Merge pull request #250 from Tencent-Hunyuan/feat/memhub-multi-user
...
feat(memhub): multi-user Runtime — user_id on memory APIs and CLI, per-user Hermes profiles, hy-memory 2.0.9
2026-11-01 14:25:05 +08:00
feitianxiang
0965519e16
Merge pull request #249 from Tencent-Hunyuan/fix/hermes-injected-content
...
fix(memhub): drop Hermes gateway notices and unwrap relayed messages on import; no false 'damaged log' warning
2026-11-01 14:24:05 +08:00
feitianxiang
97d9309ca8
Merge pull request #248 from Tencent-Hunyuan/feat/sdk-2.0.9
...
fix(sdk): placeholder values never overwrite the basic profile (2.0.9)
2026-11-01 08:26:49 +08:00
feitianxiang
988667cd87
Merge pull request #224 from Tencent-Hunyuan/chore/build-36-docs
...
chore(release): bump MemHub build number to 36 for v1.0.5
2026-10-28 22:07:22 +08:00
feitianxiang
af25f51f9e
Merge pull request #223 from Tencent-Hunyuan/docs/release-1.0.5
...
docs: prepare MemHub v1.0.5 (news, build 35, macOS arm64 only)
2026-10-28 21:59:59 +08:00
feitianxiang
4cb33b7108
Merge pull request #217 from Tencent-Hunyuan/fix/memory-source-filter-sync
...
fix(memhub): 同步记忆筛选范围并优化首次启动状态
2026-10-28 20:32:24 +08:00
feitianxiang
5f78d67363
Merge pull request #221 from Tencent-Hunyuan/fix/linux-lifecycle-queue-check
...
fix(memhub): Linux lifecycle checks the write queue, not recall without a model
2026-10-28 20:17:45 +08:00
feitianxiang
8e56bf7a37
Merge pull request #218 from Tencent-Hunyuan/fix/sdk-client-race-and-unknown-jobs
...
fix(memhub): serialize SDK client startup, log failure frames, settle pre-upgrade unknown writes
2026-10-28 19:49:24 +08:00
feitianxiang
42c0ca1d14
Merge pull request #215 from Tencent-Hunyuan/feat/memhub-sdk-logs-quiet
...
feat(memhub): HY pipeline log off, SDK text logs only in hy_memory.log (hy-memory 2.0.8)
2026-10-28 15:46:40 +08:00
feitianxiang
3387247842
Merge pull request #214 from Tencent-Hunyuan/feat/sdk-pipeline-log-switch
...
feat(sdk): switch to turn the pipeline log off, keep conversation text out of INFO logs (2.0.8)
2026-10-28 15:34:41 +08:00
feitianxiang
8277a72aea
Merge pull request #206 from rahul05ranjan/fix/runtime-stale-lease-boot-clean
...
fix(runtime): 清理早于本次开机的服务锁，异常关机后不再崩溃循环
2026-10-28 14:42:46 +08:00
feitianxiang
5eed0f81fb
Merge pull request #200 from u7-u7/fix/literal-special-token-counting
...
fix(tokens): count literal special-token spellings as ordinary text
2026-10-28 14:36:16 +08:00
feitianxiang
59ca5afca6
Merge pull request #199 from u7-u7/fix/guarded-thread-context
...
fix(async): preserve context in guarded thread calls
2026-10-28 14:36:11 +08:00
feitianxiang
2648012ad5
Merge pull request #203 from u7-u7/fix/migration-log-write-order
...
fix(memhub): preserve migration log write order
2026-10-28 14:36:05 +08:00
feitianxiang
6ba49bf112
Merge pull request #209 from Tencent-Hunyuan/fix/memory-map-node-navigation
...
fix(memhub): open source memories from map nodes
2026-10-28 14:10:28 +08:00
feitianxiang
cf410aa9cf
Merge pull request #210 from Tencent-Hunyuan/fix/harness-refresh-on-upgrade
...
fix(memhub): refresh harness skills and MCP entry after an upgrade
2026-10-28 14:08:01 +08:00
feitianxiang
1086f4d582
Merge pull request #208 from Tencent-Hunyuan/feat/memhub-experience-memory-at
...
fix(memhub): experiences show their conversation time (hy-memory 2.0.7)
2026-10-28 11:50:41 +08:00
feitianxiang
bdd2d9cf84
Merge pull request #207 from Tencent-Hunyuan/feat/sdk-coding-memory-at
...
feat(sdk): coding and working experiences keep their conversation time (2.0.7)
2026-10-28 11:33:29 +08:00
feitianxiang
f7961f340a
Merge pull request #198 from Tencent-Hunyuan/fix/memory-detail-layout
...
fix(memhub): 稳定记忆页并优化设置交互
2026-10-28 00:54:42 +08:00
feitianxiang
cfebae118d
Merge pull request #202 from Tencent-Hunyuan/fix/sdk-write-pipeline
...
fix(memhub): SDK write pipeline — turn grouping, thinking off, manual adds, home count
2026-10-28 00:54:05 +08:00
feitianxiang
e09fbd16f9
Merge pull request #197 from Tencent-Hunyuan/feat/cli-simplify
...
feat(memhub): simplify the CLI and drop the notes concept
2026-10-27 20:21:20 +08:00
feitianxiang
a0221bc434
Merge pull request #196 from Tencent-Hunyuan/fix/migration-failure-freeze
...
fix(memhub): restore and freeze the home when an upgrade migration fails
2026-10-27 19:18:19 +08:00
feitianxiang
b43c967baa
Merge pull request #195 from Tencent-Hunyuan/feat/memory-panel-all-scenes
...
feat(memhub): memory panel lists and searches chat, working and coding
2026-10-27 18:04:32 +08:00
feitianxiang
0b775d7d61
Merge pull request #194 from Tencent-Hunyuan/release/hy-memory-2.0.6
...
release: hy-memory 2.0.6
2026-10-27 15:59:35 +08:00
feitianxiang
2102a298d4
Merge pull request #193 from Tencent-Hunyuan/fix/sdk-near-dup-gate
...
fix(sdk): gate near-duplicate demotion with the MMR top-k threshold
2026-10-27 15:42:38 +08:00
feitianxiang
1374ac26b7
Merge pull request #192 from Tencent-Hunyuan/feat/sdk-search-scenes
...
feat(sdk): rerank switch, size-gated MMR, scene="all" search and list, productivity edit/delete
2026-10-27 15:22:41 +08:00
feitianxiang
49ad1b3086
Merge pull request #185 from Tencent-Hunyuan/fix/migration-chroma-settings
...
fix: list and count every memory scene, and migrate without a Chroma clash
2026-10-27 14:35:45 +08:00
feitianxiang
e5c8cc3819
Merge pull request #190 from Tencent-Hunyuan/feat/mcp-three-tools
...
feat(memhub): cut the MCP server to recall, write and diagnostic trace
2026-10-27 14:13:53 +08:00
feitianxiang
5a5650fc94
Merge pull request #182 from Tencent-Hunyuan/chore/bump-build-30
...
chore(release): bump MemHub build number to 30
2026-10-27 09:21:00 +08:00
feitianxiang
0dd2bcd6b1
Merge pull request #181 from Tencent-Hunyuan/fix/core-ready-before-recovery
...
fix(memhub): make Core ready before it resumes unfinished work
2026-10-27 09:13:25 +08:00
feitianxiang
0abf87d4b6
Merge pull request #180 from Tencent-Hunyuan/fix/memhub-sdk-only-migration-guard
...
fix(memhub): serve SDK only and stop on an unfinished upgrade migration
2026-10-26 22:34:53 +08:00
feitianxiang
e9cce69650
Merge pull request #179 from Tencent-Hunyuan/fix/memory-search-button
...
fix(memhub): search on submit and set SDK recall limits apart
2026-10-26 16:08:51 +08:00
feitianxiang
1c0cbf6d05
Merge pull request #176 from Tencent-Hunyuan/fix/memory-page-sdk-consistency
...
fix(memhub): make the SDK memory page one consistent page
2026-10-26 09:10:22 +08:00
feitianxiang
fcb1bcd866
Merge pull request #174 from Tencent-Hunyuan/chore/bump-build-22
...
chore(release): bump MemHub build number to 22
2026-10-25 15:32:06 +08:00
feitianxiang
a4b0969248
Merge pull request #173 from Tencent-Hunyuan/fix/memory-library-build23-acceptance
...
fix(memhub): repair Build 23 memory library issues
2026-10-25 15:28:12 +08:00
feitianxiang
99e3c8d9d7
Merge pull request #172 from Tencent-Hunyuan/fix/sdk-memory-native-detail
...
fix(memhub): reuse the memory map for SDK-native details
2026-10-25 09:54:03 +08:00
feitianxiang
0aeb3e8760
Merge pull request #167 from Tencent-Hunyuan/fix/simplify-sdk-migration
...
fix(memhub): migrate to SDK authority without copying memory
2026-10-24 23:18:32 +08:00
feitianxiang
70aaa7be85
Merge pull request #170 from Tencent-Hunyuan/fix/upgrade-runtime-and-sdk-memory-map
...
fix(memhub): install the bundled Runtime on upgrade and show SDK memories
2026-10-24 23:12:03 +08:00
feitianxiang
0c36a6c960
Merge pull request #168 from Tencent-Hunyuan/feat/single-sdk-migrate-command
...
feat(memhub): migrate every install with one sdk-migrate command
2026-10-24 11:47:25 +08:00
feitianxiang
083b837f32
Merge pull request #154 from Tencent-Hunyuan/fix/sdk-migration-turns-and-coding
...
fix(memhub): harden SDK migration inventory and fencing
2026-10-23 17:37:15 +08:00
feitianxiang
3b44d8381c
Merge pull request #153 from Tencent-Hunyuan/fix/sdk-migration-turns-and-coding
...
fix(memhub): repair legacy SDK authority migration
2026-10-23 17:27:39 +08:00
feitianxiang
4c311b2e93
Merge pull request #149 from Tencent-Hunyuan/fix/memhub-1.0.5-release-doc-test
...
test(memhub): update 1.0.5 release doc expectations
2026-10-22 21:31:04 +08:00
feitianxiang
0a8e43e570
Merge pull request #148 from Tencent-Hunyuan/release/memhub-1.0.5
...
release(memhub): bump to 1.0.5 and pin hy-memory 2.0.5
2026-10-22 21:22:40 +08:00
feitianxiang
d08e070816
Merge pull request #147 from Tencent-Hunyuan/release/hy-memory-2.0.5
...
release: hy-memory 2.0.5
2026-10-22 20:58:49 +08:00
26c33ee40f
refactor(sdk): internalize legacy data migration ( #146 )
...
* refactor(sdk): keep legacy migration out of client APIs
Co-authored-by: Cursor <cursoragent@cursor.com >
* fix(migration): stage SDK data before cutover
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-22 20:49:23 +08:00
d926844173
refactor(memhub): remove private Hy Memory bridge ( #145 )
...
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: kristinhe <kristinhe@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-22 20:02:27 +08:00
de4292ed4e
feat(hermes): add managed native prefetch provider ( #143 )
...
* feat: add native Hermes memory provider
* feat(hermes): support scoped private CA trust
* feat(hermes): bound write timeout separately from reads
* fix(hermes): support truststore scoped CA snapshots
* feat(hermes): add managed prefetch provider
Co-authored-by: kindong <727777477@qq.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: kindong <727777477@qq.com >
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-22 16:51:15 +08:00
366befc2aa
feat(memhub): add extensible Chinese, English, and Spanish UI ( #142 )
...
* Add Spanish localization and persistent GUI language preferences to MemHub
Co-authored-by: Cursor <cursoragent@cursor.com >
* Extend MemHub Spanish localization to the remaining GUI screens
Move the copy of Applications, memory space and map, matters, workbench,
call history, handoff and arrival panels into shared contracts catalogs.
- Merge domain catalogs without silent overrides: a key shared by two
catalogs must have identical Chinese and Spanish texts.
- Add Spanish singular variants (`.one`) and count helpers that keep the
Chinese numbers exactly as they were rendered before.
- Build the Spanish connection prompt from a catalog template; MCP and
shell snippets are inserted verbatim, the Chinese prompt is unchanged.
- Word host failures at render time so a language switch re-words them.
- Unknown text, names, paths and error codes pass through untranslated;
unknown graph node types and edge origins keep their previous labels.
- Add catalog parity, interpolation, adapter and regression tests.
* feat(memhub): add Chinese English Spanish localization
Co-authored-by: Nicolas Pirello <nicoalepirello@gmail.com >
Co-authored-by: kristinhe <kristinhe@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
* style(memhub): satisfy Node workspace lint
Co-authored-by: Cursor <cursoragent@cursor.com >
* test(memhub): align localization fixtures with current keys
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: Nicolas Pirello <nicoalepirello@gmail.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: kristinhe <kristinhe@tencent.com >
2026-10-22 16:36:37 +08:00
a4c8c7e5fb
docs: finalize public runtime and migration contracts ( #141 )
...
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-22 15:38:41 +08:00
b7d9f0e0f4
docs: remove internal handoff and stale architecture notes ( #140 )
...
* chore: keep HANDSOFF local only
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs: remove internal notes and stale architecture markup
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-22 15:34:46 +08:00
6d1d29e16d
refactor(memhub): move memory authority to SDK ( #132 )
...
* docs(memhub): define SDK-owned memory architecture
Co-authored-by: Cursor <cursoragent@cursor.com >
* test(memhub): freeze legacy memory authority debt
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(memhub): define SDK data-plane migration contract
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): add SDK cutover migration ledger
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): create verified SDK migration bundles
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): expose offline migration bundle command
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(memhub): keep durable write scheduling in Core
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): delegate SDK backup to public API
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): restore complete SDK migration bundles
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): route SDK snapshots through bridge
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): persist SDK cutover state
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): add SDK-native memory projection
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): recover writes from SDK outcomes
Co-authored-by: Cursor <cursoragent@cursor.com >
* chore(memhub): pin SDK authority dev preview
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): classify legacy and SDK memory inventories
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): trust SDK physical delete receipts
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): execute classified memory migrations
Co-authored-by: Cursor <cursoragent@cursor.com >
* chore(memhub): pin exact migration SDK dev preview
Co-authored-by: Cursor <cursoragent@cursor.com >
* fix(memhub): preserve source SDK version in migration bundle
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): use one schema-driven legacy migration
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): migrate Evidence into SDK raw
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): run resumable offline SDK backfill
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): expose offline SDK backfill command
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): expose SDK-native memory APIs
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(web): add SDK-native memory client
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): migrate recall scopes by memory id
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): remove fabricated per-memory harness ACL
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(web): remove legacy per-memory recall controls
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(memhub): use SDK source filters for future harness policy
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): remove per-memory scope mutation API
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(web): remove dead recall scope component
Co-authored-by: Cursor <cursoragent@cursor.com >
* chore(memhub): pin raw-search SDK dev preview
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): search SDK raw through existing API
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(memhub): add honest SDK-native Context v2
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): switch library to sdk authority
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): cut host context over to sdk v2
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(memhub): run migration during app startup
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): add sdk raw scheduling receipts
Co-authored-by: Cursor <cursoragent@cursor.com >
* Revert "refactor(memhub): add sdk raw scheduling receipts"
This reverts commit d07ae91c93cursoragent@cursor.com >
* refactor(memhub): stop governance mirroring after cutover
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): fence startup migration and resume effects
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): release transient evidence after sdk success
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): gate startup authority cutover
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): run sdk migration in app loading
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): block cutover until product paths switch
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): resume sdk staging cleanup on startup
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): add sdk-native ask v2
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): migrate matters to sdk references
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): hydrate matters from sdk references
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): switch matter reads to sdk references
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): queue sdk matter notes
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): discover matters from sdk authority
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): retire legacy digest and boundary after cutover
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): complete sdk raw and matter cutover
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): stop legacy maintenance after cutover
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): retire remaining legacy read surfaces
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(memhub): document sdk matter and raw cutover
Co-authored-by: Cursor <cursoragent@cursor.com >
* test(runtime): allow receipt write under parallel load
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(memhub): publish sdk authority architecture
Co-authored-by: Cursor <cursoragent@cursor.com >
* test(memhub): repair acceptance gate fixtures
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(memhub): separate fresh schema from legacy cleanup
Co-authored-by: Cursor <cursoragent@cursor.com >
* refactor(memhub): finish SDK authority cutover
Co-authored-by: Cursor <cursoragent@cursor.com >
* fix(memhub): close CodeQL exposure paths
Co-authored-by: Cursor <cursoragent@cursor.com >
* fix(memhub): hash workspace identities without path expansion
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(memhub): add app migration acceptance guide
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-22 15:24:20 +08:00
feitianxiang and alvinfei
c0737acb44
Revert "feat(sdk): process existing raw records ( #136 )" ( #138 )
...
This reverts commit 0244ebe38calvinfei@tencent.com >
2026-10-22 06:49:10 +08:00
0244ebe38c
feat(sdk): process existing raw records ( #136 )
...
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 21:45:12 +08:00
e3b5a1bc9c
release: publish Hy Memory 2.0.5.dev3 ( #135 )
...
* release: prepare Hy Memory 2.0.5.dev3
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(sdk): record 2.0.5.dev3 publication
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 20:50:45 +08:00
8aeb042bb7
feat(sdk): search raw through existing search API ( #134 )
...
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 17:58:29 +08:00
635d2572c9
release: prepare Hy Memory 2.0.5.dev2 ( #133 )
...
* release: prepare Hy Memory 2.0.5.dev2
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(sdk): record 2.0.5.dev2 publication
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 17:19:09 +08:00
6113e2351e
feat(sdk): import semantic memories exactly ( #131 )
...
* feat(sdk): import semantic memories exactly
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(sdk): record exact memory migration API
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 17:05:24 +08:00
b5a8a78c69
release: prepare Hy Memory 2.0.5.dev1 ( #130 )
...
* release: prepare Hy Memory 2.0.5.dev1
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(sdk): record 2.0.5.dev1 publication
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 17:01:12 +08:00
fd12c8f1c0
feat(sdk): add MemHub authority migration contracts ( #129 )
...
* feat(sdk): expose memory lifecycle status filters
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(sdk): persist idempotent write outcomes
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(sdk): add explicit exact raw APIs
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(sdk): snapshot quiesced local data plane
Co-authored-by: Cursor <cursoragent@cursor.com >
* feat(sdk): restore quiesced local data plane
Co-authored-by: Cursor <cursoragent@cursor.com >
* fix(sdk): report incomplete physical deletes
Co-authored-by: Cursor <cursoragent@cursor.com >
* docs(sdk): record authority migration contracts
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 16:43:25 +08:00
e089d4b800
fix(coding): default-off search chain_block (diff) injection ( #128 )
...
Medium/heavy coding users pay large prompt tokens for the version-evolution
diff block; make MEMORY_CODING_CHAIN_BLOCK_ENABLED opt-in (default false).
Write-side observe is unchanged; HCM_BLOCK_OFF remains an advanced kill switch.
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 14:46:23 +08:00
b370663870
fix(ci): locate nested Linux release artifacts ( #36 )
...
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 14:14:24 +08:00
8ea6996859
chore(core): retire memory evidence freshness on read paths ( #127 )
...
* chore(core): retire memory evidence freshness on read paths
Stop re-validating support_refs against live evidence during library,
context, and call-reference reads. The check hid memories when evidence
changed but caused 503s and segment-key breakage; proofs may still be
written but are ignored until a later teardown removes them.
Co-authored-by: Cursor <cursoragent@cursor.com >
* test(docs): align install-docs pin with hy-memory 2.0.4
CI memhub failed because README already documents 2.0.4 / v1.0.4 while
the install-docs contract test still expected 2.0.3 / v1.0.3.
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 14:11:57 +08:00
2cea39da4f
fix(core): drop native store write gate on HY search ( #126 )
...
MemHub expects concurrent native reads during precipitation. The
process-local read slot mostly existed for a Chroma/GIL fail-fast and
was never held by writers on main; activating it (as in #98 ) would only
push recall onto weaker fallbacks. Remove the lock module and the
hy_memory_store_busy path instead.
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 12:15:26 +08:00
3a74a33456
feat(memhub): pair unlisted harnesses with one-time codes ( #124 )
...
* feat: pair custom harnesses with one-time codes
* fix(memhub): satisfy biome format checks for custom harness pairing
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-21 11:36:55 +08:00
0afe787c95
fix: stop authority-invalid HY retries ( #78 )
...
* fix: prevent authority-invalid HY retries
Co-authored-by: Cursor <cursoragent@cursor.com >
* fix: align coding authority with conversations
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-19 15:30:32 +08:00
67ae43f920
fix(memhub): repair visibility dirty trigger conflicts ( #77 )
...
* fix(memhub): repair visibility link trigger
Co-authored-by: Cursor <cursoragent@cursor.com >
* fix(memhub): make trigger upgrade atomic
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-19 15:30:11 +08:00
3ce89ea560
fix(sdk): serialize sqlite connection lifecycle ( #73 )
...
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-19 13:02:22 +08:00
7b7fd217a8
fix(sdk): clarify metadata and allow opt-in raw get ( #65 )
...
Co-authored-by: alvinfei <alvinfei@tencent.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-18 22:30:23 +08:00
feitianxiang
