Back to corpus
proposalexperiment writeup candidatescore 24

DLM Reorganization - Progress Status

``` packages/dlm/ ├── response/ │ ├── techniques/ ✅ RENAMED (was vangaurd/) │ │ ├── fitness/ │ │ ├── synth/ │ │ └── word_weaver/ │ ├── system.py │ ├── links.py │ ├── cohort.py │ └── ... (other files) │ ├── inference/ │ ├── utils/ ✅ NEW SUBFOLDER │ │ ├── __init__.py │ │ └── file.py (Element, SimpleDirectoryReader, generate_id) │ ├── prompts/ ✅ NEW SUBFOLDER │ │ ├── __init__.py │ │ ├── manager.py (PromptManager) │ │ └── templates.py (SYSTEM_PROMPT_* constants) │ ├── management/ ✅ NEW SUBFOLDER │ │ ├── __init__.py │ │

Open full HTMLDownload HTMLCorpus JSONPaper schema

Full HTML reader

Read the full artifact

Open in new tab

Extracted abstract or opening context

### Phase 1: Response Module - Step 1.1 ✅ **Date**: 2025-12-08 **Action**: Renamed `vangaurd/` → `techniques/` #### Changes Made: 1. ✅ Renamed directory: `response/vangaurd/` → `response/techniques/` 2. ✅ Updated `response/__init__.py`: Changed import from `.vangaurd` to `.techniques` 3. ✅ Updated `response/cohort.py`: Changed all references from `vangaurd` to `techniques` #### Files Modified: - `packages/dlm/response/` - directory renamed - `packages/dlm/response/__init__.py` - import updated - `packages/dlm/response/cohort.py` - all references updated ### Phase 2: Inference Module - Step 2.1 ✅ **Date**: 2025-12-08 **Action**: Created `inference/utils/` subfolder for file operations #### Changes Made: 1. ✅ Created directory: `inference/utils/` 2. ✅ Moved `file_manager.py` → `utils/file.py` 3. ✅ Created `utils/__init__.py` with exports for Element, SimpleDirectoryReader, generate_id 4. ✅ Kept `cloud_manager.py` at root level (depends on PromptManager, not a true util) 5. ✅ Updated all imports across 6 files: - `inference/__init__.py` - `inference/generator.py` - `inference/manager.py` - `inference/conversation_manager.py` - `inference/prompt_manager.py` - `engine/builder.py`

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.