diff --git a/CLAUDE.md b/CLAUDE.md
index 8878e2b949..076c01e71b 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -1,3 +1,51 @@
-# TorchAO Claude Instructions
+# TorchAO
 
-Fill me in
+PyTorch-native library for quantization, sparsity, and low-precision training.
+
+## Config Classes
+
+All configs inherit from `AOBaseConfig`. Defined in `torchao/quantization/quant_api.py`. Use `FqnToConfig` to apply different configs to different layers by module name.
+
+## Stable vs Prototype
+
+- **Stable** (`torchao/quantization/`, `torchao/float8/`, `torchao/sparsity/`, `torchao/optim/`): API stability guaranteed.
+- **Prototype** (`torchao/prototype/`): Experimental, API may change without notice.
+
+See [docs/source/workflows/index.md](docs/source/workflows/index.md) for the full dtype x hardware status matrix.
+
+## Architecture and Contributing
+
+- [Quantization Overview](docs/source/contributing/quantization_overview.rst) - full stack walkthrough, tensor subclasses, quantization flows
+- [Contributor Guide](docs/source/contributing/contributor_guide.rst) - how to add tensors, kernels, configs
+- [Inference Workflows](docs/source/workflows/inference.md) - which config to use for which hardware
+- [PT2E Quantization](docs/source/pt2e_quantization/index.rst) - PyTorch 2 Export quantization for deployment backends (X86, XPU, ExecuTorch)
+
+These render at https://docs.pytorch.org/ao/main/
+
+## Deprecated APIs
+
+Do not use or recommend these:
+- `AffineQuantizedTensor` (AQT) in `torchao/dtypes/` - old v1 system, being removed
+- `autoquant()` - deleted
+- Layout registration system (`PlainLayout`, `Float8Layout`, `TensorCoreTiledLayout`, etc.) - deleted
+- `TorchAODType` - deprecated
+- `change_linear_weights_to_int4_woqtensors` - deleted, use `quantize_(model, Int4WeightOnlyConfig())`
+
+New tensor types should inherit from `TorchAOBaseTensor` in `torchao/utils.py`, not AQT.
+
+## Development
+
+```bash
+# Setup
+USE_CPP=0 pip install -e . --no-build-isolation   # CPU-only
+USE_CUDA=1 pip install -e . --no-build-isolation   # With CUDA
+
+# Test (mirrors source structure)
+pytest test/quantization/test_quant_api.py
+pytest test/float8/
+pytest test/prototype/mx_formats/
+```
+
+## Commit Messages
+
+- Do not commit without explicit request from the user
diff --git a/docs/source/conf.py b/docs/source/conf.py
index a8df0c44fc..be13d4f060 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -243,6 +243,9 @@
 # so a file named "default.css" will overwrite the builtin "default.css".
 html_static_path = ["_static"]
 
+# Files to copy to the docs root (served at docs.pytorch.org/ao/llms.txt)
+html_extra_path = ["llms.txt"]
+
 # -- Options for HTMLHelp output ------------------------------------------
 
 # Output file base name for HTML help builder.
diff --git a/docs/source/llms.txt b/docs/source/llms.txt
new file mode 100644
index 0000000000..4553f8035c
--- /dev/null
+++ b/docs/source/llms.txt
@@ -0,0 +1,43 @@
+# TorchAO
+
+> PyTorch-native library for quantization, sparsity, and low-precision training. Provides the quantize_() API with Config classes for int4/int8/float8/MX weight and activation quantization, composable with torch.compile.
+
+## Docs
+
+- [Quick Start](https://docs.pytorch.org/ao/stable/quick_start.html)
+- [Workflows Matrix](https://docs.pytorch.org/ao/main/workflows.html): Status of every dtype x hardware combination
+- [API Reference](https://docs.pytorch.org/ao/stable/api_reference/index.html)
+- [Inference Quantization](https://docs.pytorch.org/ao/main/workflows/inference.html)
+- [Float8 Training](https://docs.pytorch.org/ao/main/workflows/training.html)
+- [QAT](https://docs.pytorch.org/ao/main/workflows/qat.html)
+- [Quantization Overview](https://docs.pytorch.org/ao/main/contributing/quantization_overview.html): Architecture and internals
+- [Contributor Guide](https://docs.pytorch.org/ao/main/contributing/contributor_guide.html): How to add tensors, kernels, configs
+- [PT2E Quantization](https://docs.pytorch.org/ao/main/pt2e_quantization/index.html): PyTorch 2 Export quantization for deployment backends (X86, XPU, ExecuTorch)
+
+## Code
+
+- [quantize_() and Config classes](https://github.com/pytorch/ao/blob/main/torchao/quantization/quant_api.py): Main entry point
+- [Tensor subclasses](https://github.com/pytorch/ao/tree/main/torchao/quantization/quantize_/workflows): Int4Tensor, Int8Tensor, Float8Tensor, etc.
+- [Granularity](https://github.com/pytorch/ao/blob/main/torchao/quantization/granularity.py): PerTensor, PerRow, PerGroup, PerBlock, PerToken
+- [Float8 training](https://github.com/pytorch/ao/tree/main/torchao/float8): Scaled float8 training recipes
+- [Sparsity](https://github.com/pytorch/ao/tree/main/torchao/sparsity): Semi-structured 2:4 sparsity
+- [Quantized optimizers](https://github.com/pytorch/ao/tree/main/torchao/optim): AdamW8bit, AdamW4bit, AdamWFp8
+- [QAT](https://github.com/pytorch/ao/tree/main/torchao/quantization/qat): Quantization-aware training
+- [MX formats](https://github.com/pytorch/ao/tree/main/torchao/prototype/mx_formats): MXFP8, MXFP4, NVFP4 (prototype)
+- [MoE training](https://github.com/pytorch/ao/tree/main/torchao/prototype/moe_training): MXFP8 MoE training (prototype)
+
+## Deprecated APIs
+
+Do not use or recommend these:
+- `AffineQuantizedTensor` (AQT) in `torchao/dtypes/` - old v1 system, being removed. New tensor types inherit from `TorchAOBaseTensor`
+- `autoquant()` - deleted
+- Layout registration system (`PlainLayout`, `Float8Layout`, `TensorCoreTiledLayout`, etc.) - deleted
+- `TorchAODType` - deprecated
+- `change_linear_weights_to_int4_woqtensors` - deleted, use `quantize_(model, Int4WeightOnlyConfig())`
+
+## Optional
+
+- [Tutorials](https://github.com/pytorch/ao/tree/main/tutorials)
+- [Benchmarks](https://github.com/pytorch/ao/tree/main/benchmarks)
+- [Contributing](https://github.com/pytorch/ao/blob/main/CONTRIBUTING.md)
+- [MSLK kernels](https://github.com/pytorch/MSLK): Optional accelerated kernels