Mohamed Diomande

Full HTML reader

Read the full artifact

Extracted abstract or opening context

I've successfully cleaned up the root folder and created a globally accessible command-line tool for the IRCP hierarchical semantic search engine. ### **🗂️ Files Moved and Organized:** - **7 analysis tools** → `analysis_tools/` - **5 precomputation scripts** → `precomputation_scripts/` - **6 demo scripts** → `demo_scripts/` - **4 main tools** → `tools/` - **2 log files** → `logs/` - **Database files** → `databases/` - **Utility scripts** → `scripts/` ### **🔧 Global Tool Features** - **Works from any directory** in the terminal - **Automatic environment activation** (uses tpo_env) - **Full argument support** with all search options - **Interactive mode** available globally - **Database auto-detection** from project location ### **✅ Database Status** - **853 conversations processed** (96% of 891 total) - **16,858 messages with embeddings and coordinates** - **100% coordinate success rate** - **Two databases available** with full functionality ### **✅ Clean Architecture** - **Organized file structure** with logical groupings - **No loose files** in root directory - **Clear separation** of tools, scripts, and data - **Professional project layout**

Promotion decision

What has to happen next

Attach run IDs, datasets, metrics, and reproduction commands.

Why this is not always a full paper yet

Corpus pages are public-safe readers for discovered workspace artifacts. They are not automatically final papers. A corpus item becomes a polished paper only after the editable source, evidence checkpoints, references, figures, render path, and release status are attached through the paper schema.