fix: pin new capture to top of Just Added feed regardless of server clock skew

fix: delete button in detail view, persist AgentGuess IDs across refresh

fix: exclude failed captures from graph view

feat: show error messages when AI model unavailable in BRIEF and ASK

fix: BRIEF today timezone, exclude failed captures from digest

feat: show error toast when capture processing fails

Update Readme.md for submission

feat: add field notes blog post, badge tags (off-brand, off-the-grid, field-notes)

feat: add demo video and LinkedIn post links to README

fix: recover AgentGuess on page refresh

fix: delete button on failed capture cards

fix: PATH A embed uses sentence-transformers directly, not LM Studio API

fix: PATH A embed model switched from nomic to BGE-base

fix: rank-based graph edges (top 2 + 0.40 floor) instead of threshold

feat: better ASK suggestions matching seed content

chore: pin Space to profile

fix: short_description under 60 chars, broaden from engineers to everyone

fix: backfill only updates related_ids (not wipe claims/recall), raise BGE threshold to 0.62, README hackathon tags

feat: engineering seed, BGE-base embed, FORCE_RESEED startup flag

refactor: abstract VL chat template into _make_vl_inputs, model-agnostic

fix: Qwen2.5-VL chat template, skip embed on empty summary

fix: VL device detection, failed-capture error state, HF_HOME=/data/.cache

fix: lazy VL import, bump transformers>=4.49

fix: review queue requires recall_question, delete only removes UI on success

fix: embed device=cpu, VL model class, file input reset

fix: rebuild static with _app assets dir, remove old assets/

fix: disable Gradio SSR Node proxy, broaden GET / route filter

fix: remove Gradio GET / route so React SPA serves at root

fix: use gr.Blocks + demo.launch(prevent_thread_lock) for ZeroGPU compatibility

fix: rewrite app.py with gradio.Server for ZeroGPU compatibility

fix: bare @spaces.GPU in app.py, no try/except

fix: add @spaces.GPU directly in app.py for ZeroGPU startup detection

fix: expose @spaces.GPU functions in app.py for ZeroGPU detection, drop bitsandbytes

fix: auto-disable LM Studio on HF Spaces via SPACE_ID detection

feat: sync all features from local-guide — BRIEF synthesis, REVIEW, weekly thread, unified lm.py, new frontend build

fix: lower similarity threshold to 0.50 for MiniLM, backfill on every restart

fix: use BitsAndBytesConfig for 4-bit quant, add torchvision for vision model

fix: run uvicorn unconditionally so HF Spaces keeps the process alive

fix: export FastAPI app directly, drop gr.mount_gradio_app to stop port conflict

fix: remove manual uvicorn.run — HF Spaces starts server automatically

fix: prevent Gradio SDK from intercepting root path

feat: add backend, frontend, seed data, and HF Spaces wiring

initial commit