SkillOpt v0.1.0: initial release

- Skill optimization framework with training loop analogy
- 11 benchmarks, 4 model backends (Azure OpenAI, Claude, Codex, Qwen)
- WebUI for browser-based training control
- Pluggable architecture for extending benchmarks and backends

.gitignore

@@ -0,0 +1,41 @@
+__pycache__/
+*.pyc
+*.egg-info/
+build/
+dist/
+site/
+
+data/
+outputs/
+logs/
+external/
+
+/BabyVision/
+/MMRB/
+/SpreadsheetBench/
+/dl4ir-searchQA/
+
+configs/local/
+configs/**/*.local.yaml
+*.local.md
+*.secret.md
+*.bak
+
+.env
+.secrets/
+.codex_azure*/
+
+# Internal docs (not for open-source release)
+docs/ablation_plan.md
+docs/ablation_paper_tables.md
+docs/ablation_paper_tables.html
+docs/experiment_commands.md
+docs/slow_update_flowchart.md
+docs/session_memory.md
+docs/harness_fresh_machine_handoff.md
+docs/harness_monitoring_memory.md
+docs/harness_reproduction_secrets.secret.md
+docs/reflact_conda_env_export.yml
+docs/reflact_overview.html
+docs/render_ablation_paper_tables.py
+docs/让*