wafer-cli 0.2.52__tar.gz → 0.2.54__tar.gz

{wafer_cli-0.2.52 → wafer_cli-0.2.54}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wafer-cli
-Version: 0.2.52
+Version: 0.2.54
 Summary: CLI for running GPU workloads, managing remote workspaces, and evaluating/optimizing kernels
 Requires-Python: >=3.11
 Description-Content-Type: text/markdown
@@ -67,11 +67,11 @@ Create and manage persistent GPU environments.
 - `B200` - NVIDIA Blackwell B200 (180GB HBM3e, CUDA) - default
 
 ```bash
-wafer workspaces list
-wafer workspaces create my-workspace --gpu B200 --wait   # NVIDIA B200
-wafer workspaces create amd-dev --gpu MI300X             # AMD MI300X
-wafer workspaces ssh <workspace-id>
-wafer workspaces delete <workspace-id>
+wafer target workspace list
+wafer target workspace create my-workspace --gpu B200 --wait   # NVIDIA B200
+wafer target workspace create amd-dev --gpu MI300X             # AMD MI300X
+wafer target workspace ssh <workspace-id>
+wafer target workspace delete <workspace-id>
 ```
 
 ### `wafer agent`
@@ -83,17 +83,17 @@ wafer agent "What is TMEM in CuTeDSL?"
 wafer agent -s "optimize this kernel" < kernel.py
 ```
 
-### `wafer evaluate`
+### `wafer tool eval`
 
 Evaluate kernel correctness and performance against a reference implementation.
 
 **Functional format** (default):
 ```bash
 # Generate template files
-wafer evaluate make-template ./my-kernel
+wafer tool eval make-template ./my-kernel
 
 # Run evaluation
-wafer evaluate --impl kernel.py --reference ref.py --test-cases tests.json --benchmark
+wafer tool eval gpumode --impl kernel.py --reference ref.py --test-cases tests.json --benchmark
 ```
 
 The implementation must define `custom_kernel(inputs)`, the reference must define `ref_kernel(inputs)` and `generate_input(**params)`.
@@ -101,64 +101,40 @@ The implementation must define `custom_kernel(inputs)`, the reference must defin
 **KernelBench format** (ModelNew class):
 ```bash
 # Extract a KernelBench problem as template
-wafer evaluate kernelbench make-template level1/1
+wafer tool eval kernelbench make-template level1/1
 
 # Run evaluation
-wafer evaluate kernelbench --impl my_kernel.py --reference problem.py --benchmark
+wafer tool eval kernelbench --impl my_kernel.py --reference problem.py --benchmark
 ```
 
 The implementation must define `class ModelNew(nn.Module)`, the reference must define `class Model`, `get_inputs()`, and `get_init_inputs()`.
 
-### `wafer wevin -t ask-docs`
+### `wafer agent -t ask-docs`
 
-Query GPU documentation using the docs template.
+Query GPU documentation using the docs template. Uses the `ask_docs` tool to search wafer's documentation corpus via the API.
 
 ```bash
-wafer wevin -t ask-docs --json -s "What causes bank conflicts in shared memory?"
-```
-
-### `wafer corpus`
-
-Download documentation to local filesystem for agents to search.
-
-```bash
-wafer corpus list
-wafer corpus download cuda-programming-guide
+wafer agent -t ask-docs -s "What causes bank conflicts in shared memory?"
 ```
 
 ---
 
 ## Customization
 
-### `wafer remote-run` options
+### `wafer tool eval` options
 
 ```bash
-wafer remote-run --image pytorch/pytorch:2.5.1-cuda12.4-cudnn9-devel -- python3 script.py
-wafer remote-run --require-hwc -- ncu --set full python3 bench.py   # Hardware counters for NCU
-```
-
-### `wafer evaluate` options
-
-```bash
-wafer evaluate --impl k.py --reference r.py --test-cases t.json \
+wafer tool eval gpumode --impl k.py --reference r.py --test-cases t.json \
     --target vultr-b200 \    # Specific GPU target
     --benchmark \            # Measure performance
     --profile                # Enable torch.profiler + NCU
 ```
 
-### `wafer push` for multi-command workflows
-
-```bash
-WORKSPACE=$(wafer push ./project)
-wafer remote-run --workspace-id $WORKSPACE -- python3 test1.py
-wafer remote-run --workspace-id $WORKSPACE -- python3 test2.py
-```
-
 ### Profile analysis
 
 ```bash
-wafer nvidia ncu analyze profile.ncu-rep
-wafer nvidia nsys analyze profile.nsys-rep
+wafer tool ncu analyze profile.ncu-rep
+wafer tool nsys analyze profile.nsys-rep
 ```
 
 ---
@@ -170,9 +146,9 @@ wafer nvidia nsys analyze profile.nsys-rep
 Bypass the API and SSH directly to your own GPUs:
 
 ```bash
-wafer targets list
-wafer targets add ./my-gpu.toml
-wafer targets default my-gpu
+wafer target config list
+wafer target config add ./my-gpu.toml
+wafer target config default my-gpu
 ```
 
 ### Defensive evaluation
@@ -180,23 +156,23 @@ wafer targets default my-gpu
 Detect evaluation hacking (stream injection, lazy evaluation, etc.):
 
 ```bash
-wafer evaluate --impl k.py --reference r.py --test-cases t.json --benchmark --defensive
+wafer tool eval gpumode --impl k.py --reference r.py --test-cases t.json --benchmark --defensive
 ```
 
 ### Other tools
 
 ```bash
-wafer perfetto <trace.json> --query "SELECT * FROM slice"   # Perfetto SQL queries
-wafer capture ./script.py                                    # Capture execution snapshot
-wafer compiler-analyze kernel.ptx                            # Analyze PTX/SASS
+wafer tool perfetto <trace.json> --query "SELECT * FROM slice"   # Perfetto SQL queries
+wafer tool capture ./script.py                                    # Capture execution snapshot
+wafer compiler-analyze kernel.ptx                                 # Analyze PTX/SASS
 ```
 
 ### ROCm profiling (AMD GPUs)
 
 ```bash
-wafer rocprof-sdk ...
-wafer rocprof-systems ...
-wafer rocprof-compute ...
+wafer tool rocprof-sdk ...
+wafer tool rocprof-systems ...
+wafer tool rocprof-compute ...
 ```
 
 ---
@@ -214,10 +190,10 @@ source ~/.zshrc  # or ~/.bashrc
 ```
 
 Now you can tab-complete:
-- Commands: `wafer eva<TAB>` → `wafer evaluate`
-- Options: `wafer evaluate --<TAB>`
-- Target names: `wafer evaluate --target v<TAB>` → `wafer evaluate --target vultr-b200`
-- File paths: `wafer evaluate --impl ./<TAB>`
+- Commands: `wafer tool ev<TAB>` → `wafer tool eval`
+- Options: `wafer tool eval --<TAB>`
+- Target names: `wafer tool eval --target v<TAB>` → `wafer tool eval --target vultr-b200`
+- File paths: `wafer tool eval gpumode --impl ./<TAB>`
 
 ---
 

{wafer_cli-0.2.52 → wafer_cli-0.2.54}/README.md

@@ -49,11 +49,11 @@ Create and manage persistent GPU environments.
 - `B200` - NVIDIA Blackwell B200 (180GB HBM3e, CUDA) - default
 
 ```bash
-wafer workspaces list
-wafer workspaces create my-workspace --gpu B200 --wait   # NVIDIA B200
-wafer workspaces create amd-dev --gpu MI300X             # AMD MI300X
-wafer workspaces ssh <workspace-id>
-wafer workspaces delete <workspace-id>
+wafer target workspace list
+wafer target workspace create my-workspace --gpu B200 --wait   # NVIDIA B200
+wafer target workspace create amd-dev --gpu MI300X             # AMD MI300X
+wafer target workspace ssh <workspace-id>
+wafer target workspace delete <workspace-id>
 ```
 
 ### `wafer agent`
@@ -65,17 +65,17 @@ wafer agent "What is TMEM in CuTeDSL?"
 wafer agent -s "optimize this kernel" < kernel.py
 ```
 
-### `wafer evaluate`
+### `wafer tool eval`
 
 Evaluate kernel correctness and performance against a reference implementation.
 
 **Functional format** (default):
 ```bash
 # Generate template files
-wafer evaluate make-template ./my-kernel
+wafer tool eval make-template ./my-kernel
 
 # Run evaluation
-wafer evaluate --impl kernel.py --reference ref.py --test-cases tests.json --benchmark
+wafer tool eval gpumode --impl kernel.py --reference ref.py --test-cases tests.json --benchmark
 ```
 
 The implementation must define `custom_kernel(inputs)`, the reference must define `ref_kernel(inputs)` and `generate_input(**params)`.
@@ -83,64 +83,40 @@ The implementation must define `custom_kernel(inputs)`, the reference must defin
 **KernelBench format** (ModelNew class):
 ```bash
 # Extract a KernelBench problem as template
-wafer evaluate kernelbench make-template level1/1
+wafer tool eval kernelbench make-template level1/1
 
 # Run evaluation
-wafer evaluate kernelbench --impl my_kernel.py --reference problem.py --benchmark
+wafer tool eval kernelbench --impl my_kernel.py --reference problem.py --benchmark
 ```
 
 The implementation must define `class ModelNew(nn.Module)`, the reference must define `class Model`, `get_inputs()`, and `get_init_inputs()`.
 
-### `wafer wevin -t ask-docs`
+### `wafer agent -t ask-docs`
 
-Query GPU documentation using the docs template.
+Query GPU documentation using the docs template. Uses the `ask_docs` tool to search wafer's documentation corpus via the API.
 
 ```bash
-wafer wevin -t ask-docs --json -s "What causes bank conflicts in shared memory?"
-```
-
-### `wafer corpus`
-
-Download documentation to local filesystem for agents to search.
-
-```bash
-wafer corpus list
-wafer corpus download cuda-programming-guide
+wafer agent -t ask-docs -s "What causes bank conflicts in shared memory?"
 ```
 
 ---
 
 ## Customization
 
-### `wafer remote-run` options
+### `wafer tool eval` options
 
 ```bash
-wafer remote-run --image pytorch/pytorch:2.5.1-cuda12.4-cudnn9-devel -- python3 script.py
-wafer remote-run --require-hwc -- ncu --set full python3 bench.py   # Hardware counters for NCU
-```
-
-### `wafer evaluate` options
-
-```bash
-wafer evaluate --impl k.py --reference r.py --test-cases t.json \
+wafer tool eval gpumode --impl k.py --reference r.py --test-cases t.json \
     --target vultr-b200 \    # Specific GPU target
     --benchmark \            # Measure performance
     --profile                # Enable torch.profiler + NCU
 ```
 
-### `wafer push` for multi-command workflows
-
-```bash
-WORKSPACE=$(wafer push ./project)
-wafer remote-run --workspace-id $WORKSPACE -- python3 test1.py
-wafer remote-run --workspace-id $WORKSPACE -- python3 test2.py
-```
-
 ### Profile analysis
 
 ```bash
-wafer nvidia ncu analyze profile.ncu-rep
-wafer nvidia nsys analyze profile.nsys-rep
+wafer tool ncu analyze profile.ncu-rep
+wafer tool nsys analyze profile.nsys-rep
 ```
 
 ---
@@ -152,9 +128,9 @@ wafer nvidia nsys analyze profile.nsys-rep
 Bypass the API and SSH directly to your own GPUs:
 
 ```bash
-wafer targets list
-wafer targets add ./my-gpu.toml
-wafer targets default my-gpu
+wafer target config list
+wafer target config add ./my-gpu.toml
+wafer target config default my-gpu
 ```
 
 ### Defensive evaluation
@@ -162,23 +138,23 @@ wafer targets default my-gpu
 Detect evaluation hacking (stream injection, lazy evaluation, etc.):
 
 ```bash
-wafer evaluate --impl k.py --reference r.py --test-cases t.json --benchmark --defensive
+wafer tool eval gpumode --impl k.py --reference r.py --test-cases t.json --benchmark --defensive
 ```
 
 ### Other tools
 
 ```bash
-wafer perfetto <trace.json> --query "SELECT * FROM slice"   # Perfetto SQL queries
-wafer capture ./script.py                                    # Capture execution snapshot
-wafer compiler-analyze kernel.ptx                            # Analyze PTX/SASS
+wafer tool perfetto <trace.json> --query "SELECT * FROM slice"   # Perfetto SQL queries
+wafer tool capture ./script.py                                    # Capture execution snapshot
+wafer compiler-analyze kernel.ptx                                 # Analyze PTX/SASS
 ```
 
 ### ROCm profiling (AMD GPUs)
 
 ```bash
-wafer rocprof-sdk ...
-wafer rocprof-systems ...
-wafer rocprof-compute ...
+wafer tool rocprof-sdk ...
+wafer tool rocprof-systems ...
+wafer tool rocprof-compute ...
 ```
 
 ---
@@ -196,10 +172,10 @@ source ~/.zshrc  # or ~/.bashrc
 ```
 
 Now you can tab-complete:
-- Commands: `wafer eva<TAB>` → `wafer evaluate`
-- Options: `wafer evaluate --<TAB>`
-- Target names: `wafer evaluate --target v<TAB>` → `wafer evaluate --target vultr-b200`
-- File paths: `wafer evaluate --impl ./<TAB>`
+- Commands: `wafer tool ev<TAB>` → `wafer tool eval`
+- Options: `wafer tool eval --<TAB>`
+- Target names: `wafer tool eval --target v<TAB>` → `wafer tool eval --target vultr-b200`
+- File paths: `wafer tool eval gpumode --impl ./<TAB>`
 
 ---
 

{wafer_cli-0.2.52 → wafer_cli-0.2.54}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "wafer-cli"
-version = "0.2.52"
+version = "0.2.54"
 description = "CLI for running GPU workloads, managing remote workspaces, and evaluating/optimizing kernels"
 readme = "README.md"
 requires-python = ">=3.11"
@@ -37,7 +37,7 @@ where = ["."]
 include = ["wafer*"]
 
 [tool.setuptools.package-data]
-wafer = ["GUIDE.md", "skills/*/SKILL.md", "corpora/**/*.md"]
+wafer = ["GUIDE.md", "skills/*/*.md"]
 
 [tool.ruff]
 line-length = 100
@@ -78,6 +78,7 @@ ignore = [
 
 [tool.ruff.lint.per-file-ignores]
 "tests/**/*.py" = ["ANN001", "ANN201", "ANN202", "ANN204"]  # Don't require type annotations in tests
+"tests/test_ncu_run_local_e2e.py" = ["PLR0915"]  # E2E test has a long sequential flow by design
 "wafer/evaluate.py" = ["PLR0915", "PLR1702", "E402", "PLW2901", "ASYNC221"]  # complex deployment flows - TODO: refactor
 "wafer/output.py" = ["ANN401"]  # Output collector uses **kwargs for flexible event data
 "wafer/autotuner.py" = ["PLR0915", "PLR1702", "B007", "B904"]  # complex sweep logic - TODO: refactor