directory-management

1# Directory Management
2 
3## Project Setup
4 
5Before any work begins, resolve the project name:
6 
71. If the project name is already known from conversation context, use it.
82. Otherwise, scan for existing `*/PLAN.md` files in the current directory. If found, ask the user if they are resuming an existing project and load that `PLAN.md` into context.
93. If no existing projects are found, recommend a ≤64-char lowercase slug based on what you know from the conversation (only `[a-z0-9-]`), or ask directly if there isn't enough context. Present the recommended name and wait for user confirmation.
10 
11Once project name is resolved:
12 
131. Create and/or use the `<experiment-name>/` directory using the confirmed name for storing all the artifacts
14 
15## Directory Structure
16 
17When working with the agent, all generated files are organized under an project directory.
18 
19```
20<project-name>/
21├── specs/  
22│   ├── PLAN.md             # Your customization plan
23├── scripts/                # Generated Python scripts
24│   ├── <project-name>_transform_fn.py
25├── notebooks/              # Generated Jupyter notebooks
26│   ├── <project-name>.ipynb
27├── manifests/              # Machine-readable outputs (JSON)
28└── agent_memory/           # Session persistence (git-ignored)
29    └── session-notes.md    # Progress, artifacts, next steps
30```