tkmind/memind
1
0
Fork 0
Code Issues Pull Requests 2 Actions Packages Projects Releases Wiki Activity
829 Commits 115 Branches 1 Tag
2fa952358736e33214b08d9b6c8bd7894e57f77a
Commit Graph

5 Commits

Author SHA1 Message Date
john 1d165bc6e3 feat(goose): complete v1.49 phase3 closeout gates and context fusion plan 
Expand Goose v1.49 smoke coverage (memory chat, portal resume, page e2e,
multiturn provider), add canary memory policy lock, refresh baselines, and
ignore one-off evidence artifacts. Document headroom-based context runtime
fusion plan; include auth, scheduled-task, and wechat intent fixes on branch.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-09-09 18:11:39 +08:00
john 8f88eec2bd Separate WeChat chat, page refine, and task flows with explicit routing.
Memind CI / Test, build, and release guards (push) Successful in 4m40s
Details 
Add channel prefixes, pin edit targets in Cursor prompts, and force fresh sessions for page refine URLs so DeepSeek/Goose chat history does not pollute page tasks.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-31 14:10:02 +08:00
john dc948e10c9 Classify WeChat page refine and MindSpace URL edits as page.generate.
Memind CI / Test, build, and release guards (push) Has been cancelled
Details 
Fixes Tang-style follow-ups like「页面还不够详细」or linked HTML update requests from falling through to Goose chat and hitting stale providers.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-31 10:44:50 +08:00
john 61ba554feb Treat WeChat travel-guide requests as page.generate and re-sync Kimi before Goose fallback.
Memind CI / Test, build, and release guards (push) Failing after 4m15s
Details 
Fixes Tang-style「做攻略」messages missing the 页面 keyword from entering Cursor, and forces provider refresh before DeepSeek/Kimi fallback after Cursor failures.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-31 10:36:04 +08:00
john 91e140d402 feat(wechat): add admin-pluggable LLM intent router separate from H5.
Memind CI / Test, build, and release guards (push) Failing after 8s
Details 
Give WeChat MP its own chat.general→page.generate LLM refinement layer with memindadm toggles, shadow mode, and canary openids so service account routing stays independent of the H5 chatIntentRouter.

Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-08-01 20:57:41 +08:00