Update magpie

Author: danielholanda

skills/magpie/.federated.json

@@ -2,8 +2,8 @@
   "source": "amd-agi-magpie",
   "repo": "AMD-AGI/Magpie",
   "ref": "main",
-  "commit": "2a9263833f71755df2a93b466cdd3a9f803fc625",
+  "commit": "02d9a7dfc3aedec0e3feadb7449f6c0c318621d5",
   "path": "skills/magpie",
   "license": "MIT",
-  "imported_at": "2026-05-28T21:37:43Z"
+  "imported_at": "2026-06-18T20:31:36Z"
 }

skills/magpie/SKILL.md

@@ -37,7 +37,7 @@ magpie analyze path/to/kernel.hip --testcase "./run_test.sh"
 - `--no-perf`: Skip performance profiling.
 - `-o`, `--output-dir`: Output directory (default: `./results`).
 
-**Config template (single kernel):** Use `kernel:` with `id`, `type`, `source_files`, `working_dir`, `testcase_command`, optional `compile_command`, `env`. See [Magpie/kernel_config.yaml.example](https://github.com/AMD-AGI/Magpie/blob/2a9263833f71755df2a93b466cdd3a9f803fc625/Magpie/kernel_config.yaml.example) and [examples/ck_gemm_add.yaml](https://github.com/AMD-AGI/Magpie/blob/2a9263833f71755df2a93b466cdd3a9f803fc625/examples/ck_gemm_add.yaml).
+**Config template (single kernel):** Use `kernel:` with `id`, `type`, `source_files`, `working_dir`, `testcase_command`, optional `compile_command`, `env`. See [Magpie/kernel_config.yaml.example](https://github.com/AMD-AGI/Magpie/blob/02d9a7dfc3aedec0e3feadb7449f6c0c318621d5/Magpie/kernel_config.yaml.example) and [examples/ck_gemm_add.yaml](https://github.com/AMD-AGI/Magpie/blob/02d9a7dfc3aedec0e3feadb7449f6c0c318621d5/examples/ck_gemm_add.yaml).
 
 ## Compare (multiple kernels)
 
@@ -59,7 +59,7 @@ magpie compare kernel1.hip kernel2.hip --testcase "./run_test.sh"
 - `--baseline`: Index of baseline kernel (default: 0).
 - `--no-perf`, `-o`: Same as analyze.
 
-Example: [examples/ck_grouped_gemm_compare.yaml](https://github.com/AMD-AGI/Magpie/blob/2a9263833f71755df2a93b466cdd3a9f803fc625/examples/ck_grouped_gemm_compare.yaml).
+Example: [examples/ck_grouped_gemm_compare.yaml](https://github.com/AMD-AGI/Magpie/blob/02d9a7dfc3aedec0e3feadb7449f6c0c318621d5/examples/ck_grouped_gemm_compare.yaml).
 
 ## Benchmark (vLLM / SGLang)
 
@@ -81,7 +81,7 @@ magpie benchmark --benchmark-config examples/benchmarks/benchmark_vllm_dsr1.yaml
 - `--run-mode`: `docker` (default) or `local`.
 - `--docker-image`, `--timeout`, `-o`: Override image, timeout (seconds), output dir.
 
-Example configs: [examples/benchmarks/benchmark_vllm_dsr1.yaml](https://github.com/AMD-AGI/Magpie/blob/2a9263833f71755df2a93b466cdd3a9f803fc625/examples/benchmarks/benchmark_vllm_dsr1.yaml), [docs/benchmark.md](https://github.com/AMD-AGI/Magpie/blob/2a9263833f71755df2a93b466cdd3a9f803fc625/docs/benchmark.md).
+Example configs: [examples/benchmarks/benchmark_vllm_dsr1.yaml](https://github.com/AMD-AGI/Magpie/blob/02d9a7dfc3aedec0e3feadb7449f6c0c318621d5/examples/benchmarks/benchmark_vllm_dsr1.yaml), [docs/how-to/benchmark.md](https://github.com/AMD-AGI/Magpie/blob/02d9a7dfc3aedec0e3feadb7449f6c0c318621d5/docs/how-to/benchmark.md).
 
 ## Gap analysis (standalone)
 
@@ -109,7 +109,7 @@ Shows vendor, architecture, compiler, profiler. No mode required.
 
 When the user needs a kernel config file:
 
-1. Emit YAML matching the structure in [Magpie/kernel_config.yaml.example](https://github.com/AMD-AGI/Magpie/blob/2a9263833f71755df2a93b466cdd3a9f803fc625/Magpie/kernel_config.yaml.example): `kernel:` with `id`, `type` (hip|cuda|pytorch), `source_files`, `working_dir`, `testcase_command`, and optionally `compile_command`, `env`.
+1. Emit YAML matching the structure in [Magpie/kernel_config.yaml.example](https://github.com/AMD-AGI/Magpie/blob/02d9a7dfc3aedec0e3feadb7449f6c0c318621d5/Magpie/kernel_config.yaml.example): `kernel:` with `id`, `type` (hip|cuda|pytorch), `source_files`, `working_dir`, `testcase_command`, and optionally `compile_command`, `env`.
 2. Write the file to the user's requested path (e.g. `kernel_config.yaml`).
 3. Run: `magpie analyze --kernel-config <that_file>`.
 

skills/magpie/reference.md

@@ -100,4 +100,4 @@ kernels:
 
 ## Benchmark config YAML
 
-Top-level key `benchmark:` with `framework`, `model`, `precision`, `envs`, `profiler`, `gap_analysis`, `timeout_seconds`, etc. See `examples/benchmarks/benchmark_vllm_dsr1.yaml` and `docs/benchmark.md`.
+Top-level key `benchmark:` with `framework`, `model`, `precision`, `envs`, `profiler`, `gap_analysis`, `timeout_seconds`, etc. See `examples/benchmarks/benchmark_vllm_dsr1.yaml` and `docs/how-to/benchmark.md`.

skills/magpie/skill-card.md

@@ -2,7 +2,7 @@
 
 ## Description
 
-Perform GPU kernel correctness and performance evaluation and LLM inference benchmarking with Magpie (HIP/CUDA/PyTorch kernels, vLLM/SGLang benchmarks, TraceLens, and torch-trace gap analysis).
+Performs GPU kernel correctness and performance evaluation and LLM inference benchmarking with Magpie. Analyzes single or multiple kernels (HIP/CUDA/PyTorch), compares kernel implementations, runs vLLM/SGLang benchmarks with profiling and TraceLens, and runs gap analysis on torch traces.
 
 ## Owner