Fetching metadata from the HF Docker repository... 
Upload Dockerfile
39c03a8 verified - backend Support multi-format uploads (PDF/images/Excel/Word/PPT/ODF/text) with AI text extraction
- configs Upload configs/frontend_agent_config.json
- 1.52 kB initial commit
- 19 Bytes Support multi-format uploads (PDF/images/Excel/Word/PPT/ODF/text) with AI text extraction
- 5.61 kB Upload AUTO_CONTINUE_FEATURE.md
- 36 Bytes Upload BUILD_TRIGGER
- 4.59 kB Upload Dockerfile
- 2.28 kB Upload README.md
- 1.61 kB Update error report with DeepSeek V4 Flash routing fix details
- 3.29 kB patches: agent_loop detection + auto-approve toggle
- 2.75 kB Auto-continue feature for free models (#2)
- 5.09 kB Upload patch_auto_continue.py
- 8.38 kB V10: Fix scoping - store plan in ref, useEffect triggers sendMessage
- 6.86 kB V11: Clean single-file patch for auto-continue feature
- 6.63 kB V12: Fix two-line return match + Button import format + TS arrow param type
- 6.05 kB V13: Fix SyntaxError - no \u in Python source code
- 6.62 kB V14: Simple ASCII text - build first, add Vietnamese later
- 6.72 kB V15: Fix Button detection (check Button,) + fix return statement
- 6.19 kB V16: Match exact two-line return pattern - THE final fix
- 6.67 kB V17: Perfect Python - no syntax errors, exact two-line marker
- 12.4 kB V2: Fix all TS errors for auto-continue
- 8.72 kB V3: Fix all remaining TS errors - inline auto-continue in sideChannel
- 9.44 kB V4: Complete chain - useAgentChat → SessionChat → ChatInput buttons
- 8.76 kB V5: SIMPLE - just show pause button, no auto-send
- 8.88 kB V6: Fix variable name bug + use correct references
- 9.5 kB V7: Precise patches with exact source matching
- 8.51 kB V8: Safe auto-continue via useEffect (no hoisting)
- 8.13 kB V9: SIMPLEST - CustomEvent pattern, no interface/props changes
- 3.61 kB Upload patch_chat_input.py
- 8.25 kB Localize upload error messages (VI) + force cache-bust rebuild
- 7.23 kB Upload patch_frontend.py
- 4.73 kB Support multi-file uploads (many images/files at once)
- 2.56 kB Add 3 new OpenRouter models: UPSTAGE Solar Pro 4, NVIDIA Nemotron 3.5 Lightning:free, Liquid LFM 2.5 2.6B:free
- 2.86 kB Localize upload error messages (VI) + force cache-bust rebuild
- 11.3 kB Upload patch_models.py
- 6.36 kB Support multi-file uploads (many images/files at once)
- 3.29 kB Add 3 new OpenRouter models: UPSTAGE Solar Pro 4, NVIDIA Nemotron 3.5 Lightning:free, Liquid LFM 2.5 2.6B:free
- 3.39 kB Fix OpenRouter models: add custom_llm_provider='openrouter' so litellm stops failing on non-provider org prefixes (upstage/nvidia/liquid)
- 2.74 kB Upload patch_restore_full_app.py
- 1.61 kB Upload patch_space_bunny.py
- 1.04 kB Auto-continue feature for free models (#2)
- 4.65 kB Auto-continue feature for free models (#2)
