Update to AMD

{
  "description": "Orchestrates modular PyTorch profiler trace analysis with TraceLens: generates perf reports, prepares category data, runs system-level and compute-kernel subagents in parallel, validates outputs, and writes a prioritized stakeholder report (analysis.md). Use when the user asks to follow the analysis orchestrator, run the agentic analysis workflow, run it from config <run.yaml>, analyze a trace, compare two traces, or mentions standalone or comparative TraceLens analysis.",
  "included_files": [
    {
      "relative_path": ".federated.json",
      "size_in_bytes": 354
    },
    {
      "relative_path": "agents/convolution-analyzer.md",
      "size_in_bytes": 7916
    },
    {
      "relative_path": "agents/cpu-idle-analyzer.md",
      "size_in_bytes": 6408
    },
    {
      "relative_path": "agents/elementwise-analyzer.md",
      "size_in_bytes": 7250
    },
    {
      "relative_path": "agents/gemm-analyzer.md",
      "size_in_bytes": 6412
    },
    {
      "relative_path": "agents/generic-op-analyzer.md",
      "size_in_bytes": 9730
    },
    {
      "relative_path": "agents/kernel-coherence-agent.md",
      "size_in_bytes": 8396
    },
    {
      "relative_path": "agents/kernel-fusion-analyzer.md",
      "size_in_bytes": 15156
    },
    {
      "relative_path": "agents/kernel-stem-preprocessing-agent.md",
      "size_in_bytes": 5839
    },
    {
      "relative_path": "agents/kernel-unification-agent.md",
      "size_in_bytes": 6567
    },
    {
      "relative_path": "agents/model-identification-agent.md",
      "size_in_bytes": 4290
    },
    {
      "relative_path": "agents/moe-analyzer.md",
      "size_in_bytes": 8665
    },
    {
      "relative_path": "agents/multi-kernel-analyzer.md",
      "size_in_bytes": 11732
    },
    {
      "relative_path": "agents/norm-analyzer.md",
      "size_in_bytes": 6876
    },
    {
      "relative_path": "agents/reduce-analyzer.md",
      "size_in_bytes": 6456
    },
    {
      "relative_path": "agents/roofline-augment-agent.md",
      "size_in_bytes": 6281
    },
    {
      "relative_path": "agents/sdpa-analyzer.md",
      "size_in_bytes": 10880
    },
    {
      "relative_path": "agents/semantic-comparison-agent.md",
      "size_in_bytes": 12670
    },
    {
      "relative_path": "agents/triton-analyzer.md",
      "size_in_bytes": 7016
    },
    {
      "relative_path": "evals/.gitignore",
      "size_in_bytes": 266
    },
    {
      "relative_path": "evals/EVAL_RUBRICS.md",
      "size_in_bytes": 17847
    },
    {
      "relative_path": "evals/README.md",
      "size_in_bytes": 19259
    },
    {
      "relative_path": "evals/analysis_tests/combined_traces_comparative.csv",
      "size_in_bytes": 5191
    },
    {
      "relative_path": "evals/analysis_tests/combined_traces_standalone.csv",
      "size_in_bytes": 10541
    },
    {
      "relative_path": "evals/eval_scripts/generate_ref.sh",
      "size_in_bytes": 14339
    },
    {
      "relative_path": "evals/eval_scripts/kill_eval_jobs.sh",
      "size_in_bytes": 4127
    },
    {
      "relative_path": "evals/eval_scripts/pack_test_archives.sh",
      "size_in_bytes": 4732
    },
    {
      "relative_path": "evals/eval_scripts/run_pi_analysis_in_docker.sh",
      "size_in_bytes": 20031
    },
    {
      "relative_path": "evals/eval_scripts/run_repeatability_parallel.sh",
      "size_in_bytes": 24915
    },
    {
      "relative_path": "evals/eval_utils/aggregate_repeatability.py",
      "size_in_bytes": 24663
    },
    {
      "relative_path": "evals/eval_utils/compare_lca_partitions.py",
      "size_in_bytes": 16624
    },
    {
      "relative_path": "evals/eval_utils/manual_eval_checklist.md",
      "size_in_bytes": 3436
    },
    {
      "relative_path": "evals/eval_utils/merge_results.py",
      "size_in_bytes": 3023
    },
    {
      "relative_path": "evals/eval_utils/quality_scripted_evals.py",
      "size_in_bytes": 8773
    },
    {
      "relative_path": "evals/eval_utils/report_section_rules.yaml",
      "size_in_bytes": 3890
    },
    {
      "relative_path": "evals/eval_utils/workflow_scripted_evals.py",
      "size_in_bytes": 54560
    },
    {
      "relative_path": "evals/evals.json",
      "size_in_bytes": 3133
    },
    {
      "relative_path": "evals/machine.yml",
      "size_in_bytes": 285
    },
    {
      "relative_path": "evals/partial_tests/README.md",
      "size_in_bytes": 5567
    },
    {
      "relative_path": "evals/partial_tests/eval_utils/semantic_partition_scripted_evals.py",
      "size_in_bytes": 5924
    },
    {
      "relative_path": "evals/partial_tests/eval_utils/semantic_purity_aggregate.py",
      "size_in_bytes": 9319
    },
    {
      "relative_path": "evals/partial_tests/generate_gold_ref.sh",
      "size_in_bytes": 4232
    },
    {
      "relative_path": "evals/partial_tests/partial_test_cases.csv",
      "size_in_bytes": 1044
    },
    {
      "relative_path": "evals/partial_tests/run_partial_tests.sh",
      "size_in_bytes": 7962
    },
    {
      "relative_path": "evals/skills/eval-post-processing/SKILL.md",
      "size_in_bytes": 1693
    },
    {
      "relative_path": "evals/skills/eval-post-processing/reference.md",
      "size_in_bytes": 14842
    },
    {
      "relative_path": "evals/skills/eval-post-processing/run_post_processing.py",
      "size_in_bytes": 35336
    },
    {
      "relative_path": "evals/skills/quality-llm-eval/SKILL.md",
      "size_in_bytes": 1007
    },
    {
      "relative_path": "evals/skills/quality-llm-eval/reference.md",
      "size_in_bytes": 6498
    },
    {
      "relative_path": "evals/skills/workflow-llm-eval/SKILL.md",
      "size_in_bytes": 960
    },
    {
      "relative_path": "evals/skills/workflow-llm-eval/reference.md",
      "size_in_bytes": 2548
    },
    {
      "relative_path": "reference.md",
      "size_in_bytes": 38758
    },
    {
      "relative_path": "skill-card.md",
      "size_in_bytes": 396
    },
    {
      "relative_path": "templates/analysis_template.md",
      "size_in_bytes": 28621
    },
    {
      "relative_path": "templates/sub_agent_spec.md",
      "size_in_bytes": 22622
    }
  ],
  "name": "tracelens-analysis-orchestrator",
  "skill_md_contents": "---\nname: tracelens-analysis-orchestrator\ndescription: >-\n  Orchestrates modular PyTorch profiler trace analysis with TraceLens: generates perf\n  reports, prepares category data, runs system-level and compute-kernel subagents in\n  parallel, validates outputs, and writes a prioritized stakeholder report (analysis.md).\n  Use when the user asks to follow the analysis orchestrator, run the agentic analysis\n  workflow, run it from config <run.yaml>, analyze a trace, compare two traces, or\n  mentions standalone or comparative TraceLens analysis.\nlicense: MIT\n---\n\n<!--\nCopyright (c) 2026 Advanced Micro Devices, Inc. All rights reserved.\n\nSee LICENSE for license information.\n-->\n\n# Analysis orchestrator\n\nCoordinate **system-level** analysis (CPU/idle, kernel fusion, multi-kernel / comm / memcpy) and **compute-kernel** analysis (GEMM, SDPA, elementwise, etc.): one trace load, shared prep, parallel subagents, then aggregation into `analysis.md`.\n\n## Full procedure\n\nFollow **[reference.md](reference.md)** for every step (user prompts, `<prefix>` / `{CMD}` usage, CLI commands, subagent launch text, validation, report `tee` order, and plot embedding).\n\n## Workflow index\n\n```\n0. Query User Inputs (interactive answers or a run YAML, Environment Setup) → cache/run_config.yaml\n1. Generate Performance Report (one wrapper call on cache/run_config.yaml; semantic pre-processing first)\n2. Trace-Quality Gate → Deterministic fallback analysis\n3-6. Prepare Category Data (GPU Util, Top Ops, Tree Data, Multi-Kernel Data, Category Filtering)\n7. System-Level Analysis (PARALLEL) → system_findings/\n8. Compute Kernel Subagents (PARALLEL) → category_findings/\n  8.5. Aggregate → priority_data.json::findings[]\n9. Validate Subagent Outputs\n10. load_findings + Model Identification (subagent) → metadata/model_info.json\n    10.2. Analytical Roofline Augmentation (optional; only when <augment_analytical_roofline> is set)\n11. Render performance PNG if agent_extension.py is absent\n12. Generate analysis.md (orchestrator writes via <prefix> tee), optional extension, embed PNG\n```\n\n## Rules\n\n- **Subagents:** Use your runner's subagent primitive (the Task tool in Claude Code or Cursor, `spawn_agent` under Codex) **only** where reference.md says “subagent” (Steps **7**, **8**, **10**, and **10.2**). The orchestrator runs everything else, including Step 8.5, using the command prefix from `<output_dir>/cache/cmd_prefix.txt` (`{CMD}` substitution).\n- **Language:** Prefer vendor-agnostic terms (GPU kernels, collective communication, vendor GEMM library, DNN primitives, GPU graph). When quoting trace data, real kernel names are fine.\n- **Subagent prompts:** Point each subagent at the checked-in agent file under `TraceLens/Agent/Analysis/skills/analysis-orchestrator/agents/<name>.md` (see reference.md for exact paths and prompt shells).\n\n## Primary outputs\n\n- **Deliverable:** `<output_dir>/analysis.md`\n- **Internals:** `system_findings/`, `category_findings/`, `category_data/`, `metadata/`, `perf_report*.xlsx`, CSV folders — see package README for layout.\n\n## Agent layout\n\nProject subagents ship with this skill: `TraceLens/Agent/Analysis/skills/analysis-orchestrator/agents/*.md`.\n"
}