feat: P0/P1 — Image, OCR, Translation tabs + styled headers on all tabs

feat: UI theme + voice fix + ENV.md + improvements roadmap

sec: remove API key from Gradio frontend value — SEC-1 critical fix

feat: wire STT/TTS/Nemotron/Federation/Florence2 — surface hidden services

feat: P5/P6 model swap + modal fix + doc consolidation

feat(rag): docs ingestion + UI/bus enhancements (P3 continuation)

docs: live screenshot of Ask tab working with local Ollama LLM

fix: llm.chat IndexError (lazy Ollama warm + safe _resolve_backend fallback) + chat self-send returns direct

docs: live screenshot of local node meshing with HF Space over the relay

Add all-to-all internet mesh over relay hub (P1-P3) + user-story screenshot proof

docs: add local-node<->HF-Space connection proof (screenshots + make_proof.py)

feat: WebLLM browser agent with PeerJS mesh, HybridRAG, news signals, and easter-egg ticker

docs: real semantic RAG, wired sponsor backends, multi-model contract note

feat: Nemotron Space, Modal backend, sponsor prize targeting, improvements doc

fix: asyncio.get_running_loop() across backends; expand mesh/capability/model UI docs

fix: 0 test failures; FileService; real RagService; emergency probe; chat return

test: 30 user-story E2E tests (Playwright) + 31 annotated screenshots

feat: mesh graph tab, routing trace, QR invite, remove all fakes, specialized-node tests

fix: node identity, Gradio 6 chatbot format, peer discovery; add routing+special-node docs; screenshots

feat: impl_ref gaps closed — Nemotron/OpenBMB backends, spec refs, extended UI, HOWTO, multi-node E2E

feat: Phase 2 (M14-M25, X05-X07) + Phase 3 experimental (M26-M31) + E2E tests + docs

p2, p3

Create p2_p3/readme.md

prd splitted + contracts

Create roadmap.md

Update docs/prd_v2.md

initial PRD