docs: final structural reorganization

docs: mention LoRA weights availability

docs: add training logs and update baseline source

docs: add real agent trace and sandbox image

feat: refactor Task 5 pipeline and enhance sandbox security

Update docs/HFBlogPost.md

Update docs/HFBlogPost.md

docs: add testing and verification section

feat: enhance sandbox security and update docs

docs: integrate Surgisphere, Vioxx case studies and Sycophancy research citation

docs: sync updated architecture diagram

docs: add concrete statistics and cite arXiv:2601.19100

docs: sync new combined reward and loss plots

docs: sync updated reward and loss curves

docs: add baseline vs trained comparison plot

fix: restore HF Space YAML config block

docs: sync readme, blog, and images to hf

fix: implement auto-recovery fallback for sandbox MemoryError

fix: enforce csv module and robust keyword extraction

fix: genuine LLM gen for Task 5, no hardcoding, no numpy crashes

fix: deterministic Task 5 agent - fast and reliable scoring

feat: add session audit history table to Gradio UI

feat/fix: update app.py for Task 5 without repetition penalty

fix: disable LoRA for Task 5 code gen, add repetition penalty

feat: add Task 5 NDA Data Review with Python sandbox execution

fix: change fastapi root to /api so gradio can use /

fix: pin huggingface_hub<0.26 for gradio HfFolder compat

fix: upgrade gradio to 5.x for huggingface_hub compat

feat: add Gradio demo UI with trained LoRA adapter (LFS)

feat: add Gradio demo UI with live LoRA inference and deterministic grading

fix: grader1 _type_matches bug - 2/4 flaw types could never match

refactor: PeerGuard clinical trial verification system

Revert to standard python:3.11-slim

Use specific Python base image for Docker reliability

Fix grader scores to be strictly between 0 and 1

Fix structured output format with brackets

Fix OpenAI client proxy parameter conflict

Fix Phase 2 validation: inference.py error handling

Final OpenEnv submission - all requirements

Final OpenEnv submission - all requirements met

Ready for submission: HF Router + OpenEnv validated

Add Task 4 Citation Integrity Check - perfect 1.0 baseline scor

Improve graders, add efficiency bonus, unit tests, and citation

Improve graders, add efficiency bonus, unit tests, and citation

Add OpenEnv multi-mode deployment requirements

Add pyproject.toml for OpenEnv multi-mode deployment

Fix /reset endpoint for empty body and update inference.py for hackathon compliance

fix: use Request to properly handle empty/missing POST body

fix: properly handle empty POST body on reset

add root endpoint for better UX