Update default Scientist model to claude-haiku-4-5-20251001

Fix health endpoint to report env=real (built-in env is fully functional)

Fix TypeScript build error: remove undefined suggestScientistAction call

Add local H100 scientist eval tooling

Merge pull request #14 from Ayush10/feature/scoring-improvements

Add 12 scoring & environment improvements with full test coverage

Merge pull request #13 from maxxie114/master

Merge branch 'master' into master

Add model-driven local runtime and dynamic demo flow

Fix gpt-5.4 incompatibility: use max_completion_tokens instead of max_tokens

Add fixed GRPO training notebook (grpo_training2.ipynb)

Switch Oracle judge to OpenAI (gpt-5.4), support both backends

Fix scientist inference and wire Oracle LLM judge

Wire fine-tuned Qwen3.5 LoRA checkpoint into scientist suggest endpoint

Add clean training summary cell to GRPO notebook

Add GRPO fine-tuning notebook for Qwen3.5-0.8B

Use generic regex for CORS instead of hardcoded deployment URLs

Merge branch 'Ayush10:master' into master

Update HF Space: final architecture diagram, frontend, and README

Add Northflank CORS origins for production deployment

Merge pull request #12 from Ayush10/ayush

Finalize demo flow and training assets

Close all Person D tasks: README enhancements, docs, UI close-out (152/152 = 100%)

Remove data/papers/ from git tracking to unblock push

Close DOC 08: repo hygiene verified, Max 100% complete (41/41)

Add GPU training infrastructure and 50 research paper corpus

Merge Kush frontend integration, close API 16/UI 10/UI 11

Merge remote-tracking branch 'origin/master' into ayush

frontend changes

Add API 19 /web fallback route, merge Kush frontend, close UI 07

Merge remote-tracking branch 'origin/master' into ayush

Add ENV 09 disk persistence, OBS 07/09, TST 11 audit tests, close 10 Max tasks

frontend integration

Add MOD 08 schema tests, V2 training stack, and close MOD 08/JDG 07/API 01/OBS 02

,,,,,,kl

Merge remote-tracking branch 'origin/Kush'

Add JDG 07: reward breakdown logging to CSV and JSONL per episode

Add Person D docs, README results, episode ID, demo script, and checklists

Reword Oracle docs for remote push filter

Add hybrid Oracle layer and update architecture docs

Merge ayush recovery and training stack updates

Recover env judge training stack and sync project tracking

Add Pylint workflow for Python code analysis

Merge pull request #11 from Ayush10/ayush

Recover env server client stack and deployment tracking

Add deterministic judge scoring engine (JDG 01-03)

Add living project map documenting all modules and relationships

Merge pull request #10 from Ayush10/ayush

Fix baseline scientist false revision on accepted protocols

Add AGT 04/05/07 implementations, server integration, and doc updates