A cabinet of curiosities for Bande à Bonnot experiments: weird, fun, useless, interesting, unnamed, maybe-useful, maybe-not things that deserve a place to exist before anyone decides what they are.
- Tiny prototypes and one-off scripts
- Half-ideas that need a runnable sketch
- Useless-but-fun toys
- Interesting technical spikes
- Name-finding / shape-finding explorations
- Experiments that might later graduate into their own repo
If it is already a product, library, or serious long-lived project, it probably wants its own sibling repo. If it is still strange and exploratory, put it here.
.
├── AGENTS.md # Agent/project instructions
├── docs/ # Cross-experiment docs and handoff
│ ├── HANDOFF.md
│ ├── brainstorms/
│ ├── plans/
│ ├── runbooks/
│ └── solutions/
├── experiments/ # One nested directory per experiment
│ ├── README.md
│ └── _template/
└── todos/ # Markdown todo records
slug="my-weird-thing"
cp -R experiments/_template "experiments/$slug"Then edit experiments/$slug/README.md and keep all experiment-specific files inside that directory.
A minimal experiment should answer:
- What is this?
- Why is it interesting/funny/useful/useless?
- How do I run or inspect it?
- What did we learn?
- Should it be abandoned, kept, or graduated?
Freshly scaffolded. See docs/HANDOFF.md for the live agent handoff and todos/ for next setup tasks.
MIT. See LICENSE.