← all publishers

rocm

@rocm source repo

Report this profile Claim this profile

• 123 skills
• 0 followers
• 19 hours ago last updated

1. ▌
   Update Pr Review · rocm

   Reconcile fresh review findings against existing inline comment threads on a PR. Iterates over previous Claude root comments first so that fixed issues are correctly resolved, then handles still-present issues, then identifies genuinely new findings. Never posts the same issue twice.
   0
   installs
2. ▌
   Review Rocmlir Pr · rocm

   Review a rocMLIR pull request with deep expertise in MLIR/LLVM coding standards, the Rock dialect, MIGraphX integration, kernel codegen for AMD GPUs, lit/E2E testing, and the rocMLIR CMake build. Use when asked to review a rocMLIR PR or check a rocMLIR change. Read-only; never posts to GitHub.
   0
   installs
3. ▌
   Rocm Pr Quality · rocm bundle

   Help an engineer author, review, or pre-merge-gate a pull request to a ROCm library so it is traceable, tested, and safe to merge. Use when preparing a PR, reviewing a PR, or deciding whether an approved PR is safe to merge right now, or when the user provides a GitHub PR URL or branch and asks for help with PR quality, description, testing, review, or merge-readiness. Library-agnostic base; component overlays extend it.
   0
   installs
4. ▌
   Therock Pr Quality · rocm

   TheRock build-repo supplements to the ROCm PR quality base skill. Use for TheRock PR author, review, or pre-merge gating where the change touches the superbuild, submodules/patches, artifact descriptors, or reusable CI workflows. Adds and tightens base rules; never relaxes a base MUST.
   0
   installs
5. ▌
   Copyright Check · rocm

   Copyright Header Check
   0
   installs
6. ▌
   Review Xla Pr · rocm

   Review an XLA pull request with deep expertise in HLO optimizations, GPU backend, Triton codegen, autotuner, and AMD/ROCm parity. Use when asked to review an XLA PR or check an XLA change.
   0
   installs
7. ▌
   Job Report · rocm

   Generate a factual summary report card for an HPC job from an Omnistat database using the single-shot omnistat-inspect JSON command. Use this for a quick, comprehensive snapshot of what a job did (stats, energy, health, data quality) without diagnosing why. For root-cause investigation, performance debugging, or comparing jobs, use job-analysis instead.
   0
   installs
8. ▌
   Job Analysis · rocm bundle

   Analyze an HPC job from an Omnistat database using hypothesis-driven exploration, driven by the omnistat-inspect tool. Use this to diagnose why a job behaved as it did — performance bottlenecks, hardware issues, anomalies, or comparing a degraded job against a healthy baseline. For a plain factual snapshot without investigation, use job-report instead.
   0
   installs
9. ▌
   Open Database · rocm

   Open and explore an Omnistat database using VictoriaMetrics. Use this when the user wants to view, query, or visualize collected Omnistat telemetry data from a user-mode job.
   0
   installs
10. ▌
    Atom Patterns · rocm

    Coding patterns and architecture index for the ATOM LLM inference engine
    0
    installs
11. ▌
    Capture Trace · rocm

    Capture a PyTorch profiler / kineto trace from a running ATOM server for a short benchmark window. Use when the user asks for "a trace", "profiler trace", "GPU trace", or "抓 trace" for performance investigation — what kernels ran, what's on the critical path, what's slow. Do NOT use for crashes (use debug-agent-locate-kernel) or numerical bugs (use dump-bisect-debug).
    0
    installs
12. ▌
    Dump Bisect Debug · rocm

    Locate forward numerical bugs by dumping intermediate tensors from a target implementation and a known-good reference, then bisecting layer by layer. Also covers batch-invariance bisect (the same token at any batch position should produce a bitwise-identical output, per DeepSeek V4 paper §3.3). Use when "the output is wrong but I don't know where" — model produces gibberish, degenerates, or picks the wrong token, but code review reveals nothing.
    0
    installs
13. ▌
    Run Atom Workload · rocm

    Run any ATOM workload — accuracy eval (GSM8K via lm_eval), performance benchmark, concurrency sweep, offline simple_inference, or fault repro under rocm-debug-agent. Use when the user asks to "test accuracy", "测精度", "跑 GSM8K", "跑 benchmark", "test performance", "run sweep", "repro the fault", "测一下 MTP1 精度", "跑 simple_inference" — anything that drives an ATOM workload. Encodes the canonical flow (stop → start → workload-in-bg → wait_infer_drain → stop) and the model-family env vars. Same pattern works for both server-based workloads (lm_eval / benchmark client) and offline simple_inference. Do NOT use for profiling traces (use capture-trace).
    0
    installs
14. ▌
    Debug Agent Locate Kernel · rocm

    Tool selection: debug-agent vs rocgdb
    0
    installs
15. ▌
    Deploy Mori · rocm

    Deploy and set up the MORI environment in a fresh Docker container or bare host: start the container, install ROCm dependencies, NIC userspace libraries (AINIC/Broadcom/Mellanox-NVIDIA), RDMA-core, and MORI itself. Use when the user asks to deploy MORI, install MORI in a container, set up a fresh dev environment for MORI, or prepare an AINIC / Thor2 / mlx5 box for MORI.
    0
    installs
16. ▌
    Cco Sdma API · rocm

    Write GPU kernels that move data with the CCO SDMA device API (ccoSdma: put/get/quiet/waitSignal/commit). Covers the build flags and env vars needed to turn the path on, which template arguments and coop scope to pick, how completion and signals work, and the failure modes. Use when the user is writing or debugging a kernel that calls ccoSdma, asks how to do intra-node copies with the copy engine, or hits a hang or wrong data on the SDMA path.
    0
    installs
17. ▌
    Known Issues · rocm bundle

    Known environment issues that make MORI slow or behave oddly without being MORI bugs — currently: HIP VMM peer traffic silently falling off XGMI onto PCIe/host memory on Linux kernels built without CONFIG_DMABUF_MOVE_NOTIFY / CONFIG_PCI_P2PDMA, which hits mori-cco (EPv2) but not mori-shmem (EPv1). Use when EPv2 / dispatch_combine_v2 / cco is much slower than expected, when EPv2 is far slower than EPv1 on the same box, when a2a bandwidth does not scale with the number of peers, or when the user asks whether a perf problem is a MORI bug or an environment/toolchain problem. Also: (2) on ROCm 7.0-7.2.x a clr hipMemSetAccess sub-buffer bug makes mori-cco 2-node ccoDevCommCreate fail with "hipMemSetAccess ... invalid argument"; (3) on ROCm 7.2.x clr routes uncached VMM allocations to the coarse-grained pool, making cco combine ~10-15us slower.
    0
    installs
18. ▌
    Umbp Add Backend · rocm

    Add a new storage medium to UMBP's distributed mode as a MediumBackend (HBM, SSD, CXL, a remote object store, a fake for tests). Covers the two shapes a backend can take — a PageMemorySource for anything page-addressable, or a full MediumBackend for a medium whose bytes are not directly addressable — plus registration, the heartbeat/event contract, and the failure modes that only show up under the peer service. Use when the user wants to add or debug a storage tier/medium/backend in src/umbp/distributed, mentions MediumBackend, PageBackend, PageMemorySource or BackendRegistry, or asks why a new tier's puts or gets are not working.
    0
    installs
19. ▌
    Cluster Network Topology · rocm bundle

    Discover, diagram, and diagnose the RDMA/GPU network topology of a GPU cluster (rails, NICs, GID/RoCE config, GPU<->NIC PCIe affinity), determine whether the fabric supports cross-rail communication, and localize a failure to a fabric tier (NIC / leaf / spine) using only unprivileged host-side probes. Picks up where `mori check` stops: `mori check` answers "is this host configured correctly?" and drives its peer over SSH, while this answers "what is the fabric, and which tier broke?" from inside a scheduler allocation where SSH to compute nodes is unavailable. Handles IPv4-mapped AND IPv6-ULA RoCE addressing and classifies at BOTH the IP and RDMA layers — a fabric can be IP-routable cross-rail yet rail-only for RDMA. Includes the addressing-plan decode that recovers rail/leaf/pod grouping without switch access, spine enumeration that settles shared-vs-partitioned upper tiers from an unprivileged shell, a confound register for the measurement tools themselves, and an sbatch harness for multi-node tests under S
    0
    installs
20. ▌
    Umbp Add Transfer Engine · rocm

    Add a new byte-moving path to UMBP's distributed mode as a TransferEngine (GPU copy, a second RDMA transport, a file/GDS engine, a fake for tests). Covers CanHandle pair-dispatch and why engines must be disjoint, the Plan/Submit/Wait split, bounce buffers, PeerDirectory for transports that need a handshake, and how to extend TransferRef for a non-memory endpoint. Use when the user wants to add or debug a transfer engine in src/umbp/distributed/ transfer, mentions TransferEngine, CanHandle, TransferRef, MoriIoEngine, LocalCopyEngine or CompositeTransferEngine, or asks why a transfer is "unplannable" / no engine claimed a pair.
    0
    installs
21. ▌
    Review Pr · rocm bundle

    Advisory AI code review for aiter and FlyDSL PRs. Catches perf regressions, silent correctness bugs, dispatch gate holes, and AI-generated code patterns, but never acts as a merge gate. Invoke with a PR number (optionally owner/repo#N) and, when one exists, a validation report path. Step 1 triages whether the PR changes runtime surface at all and, when it does and the PR ships a single test target, runs validate-kernel-pr itself; a PR with no runtime surface is reported N/A rather than unvalidated. That run also times the target on base and head back to back on one locked GPU, so a kernel PR's latency is measured rather than assumed. The review line stays advisory; deterministic correctness and perf results are judged only from a head-matched report.
    0
    installs
22. ▌
    Aiter Op Test · rocm

    Standard structure for aiter op_tests under op_tests/test_*.py — @benchmark + run_perftest candidate loop, a torch reference, a final markdown summary table, a __main__ guard so the module is importable, and faithful reproduction of the real model call (output buffer, layout, shapes). Use whenever writing, rewriting, or extending any aiter unit/perf test, or adding model-derived shapes (e.g. DeepSeek-V4) to an existing one.
    0
    installs
23. ▌
    Aiter Config Shape · rocm

    How to add/upload tuned config CSVs under aiter/configs (incl. model_configs/) without introducing duplicate shapes, and how to find & resolve duplicate-shape collisions. Use whenever adding a model's tuned config, merging/uploading config CSVs, editing anything under aiter/configs/**, or when a run hits "duplicate shape entries during merge".
    0
    installs
24. ▌
    Validate Kernel Pr · rocm bundle

    Reproducible validation executor for kernel PRs. Applies an explicit base-to-head patch in an isolated worktree, runs it on a verified-idle GPU, compares the same targets against base, policy-checks the test diff, and emits a head-bound validation_report.json. Missing environment evidence is INCONCLUSIVE, never PASS.
    0
    installs
25. ▌
    Opus Kernel Best Practice · rocm

    Compile-time optimization guidance for HIP/C++ kernels using opus.hpp. Use when writing or reviewing OPUS kernels, analyzing compile time, reducing template instantiation overhead, or optimizing hipcc build performance.
    0
    installs
26. ▌
    Flydsl Kernel Code Cleanup · rocm

    Modernize FlyDSL kernels: replace raw MLIR dialects (arith, scf, vector, llvm, memref, math), ArithValue, redundant fx.* wrapping, fx.Index, buffer_ops, SmemPtr/SmemAllocator, copy_atom_call/mma_atom_call (loop or single atom), and raw rocdl.mfma_* with the current fx.* surface (fx types, Python control flow, make_buffer_tensor, SharedAllocator, fx.copy/fx.gemm, make_layout_tv/make_tiled_copy TV layouts, to_llvm_ptr, arch-dispatched fx.rocdl.s_waitcnt, local @flyc.jit if/else). Also trims comments and dead code and applies the _run_compiled fast launch path. Use when reviewing, cleaning, or migrating existing kernels.
    0
    installs
27. ▌
    Opus Module Build Optimization · rocm

    Module-level JIT build-wall optimization for opus-based aiter modules. Use when an aiter JIT module's first-call build wall is a user-visible bottleneck or when adding a new module.
    0
    installs
28. ▌
    Format Code · rocm

    Format, clean up, and style-check changed files, matching the project's CI style gate. Formats Python with black + ruff and C/C++ with clang-format using the repository's .clang-format, and can also run check-only to reproduce the CI gate locally without editing files. Use when the user says "format code", "clean up code", "lint", "format before commit", "/format-code", wants to reproduce the "Check Python Code Style" or "Check C++ Code Style" CI jobs locally, is fixing a CI style failure, is about to push Python or C/C++ changes, or mentions black, ruff, or clang-format.
    0
    installs
29. ▌
    Build Flydsl · rocm

    Build and install FlyDSL (Flexible Layout Python DSL) on a remote host or Docker container. FlyDSL is a Python DSL and MLIR-based compiler stack for authoring high-performance GPU kernels with explicit layouts and tiling on AMD GPUs. Requires building LLVM/MLIR from source (~30min) then FlyDSL C++ and Python bindings (~5min). Usage: /build-flydsl [container@host]
    0
    installs
30. ▌
    API Stability · rocm

    Review a FlyDSL PR, commit, branch, kernel, or consuming module for API-stability compliance. Detect breaking changes to stable APIs, usage of unstable FlyDSL APIs, and direct upstream MLIR dialect operations. Use when asked to review API compatibility, a stable-API change, or whether a kernel/module uses only stable APIs.
    0
    installs
31. ▌
    Oob Detection · rocm

    Detect out-of-bounds memory accesses in CPU or GPU code using static interval analysis and runtime assertions/printfs. Use when investigating OOB, buffer overrun, invalid memory access, HIP/ROCm illegal address, CUDA illegal memory access, silent tensor corruption, or suspicious buffer_load/store address arithmetic.
    0
    installs
32. ▌
    Build Rocm Image · rocm

    Connect to a remote host via SSH and build a Docker image with rocprofv3, aiter, and FlyDSL. Use when user wants to build/rebuild the ROCm development image on a remote host. Usage: /build-rocm-image <hostname>
    0
    installs
33. ▌
    Lds Optimization · rocm

    Optimize LDS (Local Data Share / shared memory) access patterns in FlyDSL GPU kernels. Diagnose bank conflicts and high lgkmcnt stalls from ATT trace data, then apply swizzle or padding layouts to eliminate conflicts. Also increase the distance between LDS write and subsequent LDS read to hide LDS latency. LDS read preceded by write always requires a sync (s_waitcnt lgkmcnt or s_barrier). Use when trace analysis shows ds_read/ds_write/lgkmcnt as a bottleneck. Usage: /lds-optimization
    0
    installs
34. ▌
    Gemm Optimization · rocm

    Comprehensive guide to optimizing GEMM (General Matrix Multiply) kernels in FlyDSL on AMD CDNA GPUs. Covers tiling strategy, LDS ping-pong double-buffer, XOR bank-conflict swizzle, A/B data prefetch pipeline, 2-stage software pipelining, MFMA instruction scheduling (hot_loop_scheduler), epilogue strategies (direct store vs CShuffle), TFLOPS/bandwidth calculation, main-loop instruction count analysis, and bottleneck identification from ATT traces. Based on the production preshuffle_gemm kernel. Usage: /gemm-optimization
    0
    installs
35. ▌
    Isa Resource Diff · rocm bundle

    Detect per-kernel GPU resource regressions (VGPR, SGPR, register spills, scratch, static LDS) by diffing the final ISA before and after a change, using its `isa_resource_table.py` helper. Compile-only: needs no GPU and no profiler run, so it works on any target the compiler supports and runs in seconds. Use when asked whether a change increased register pressure, caused spilling, or hurt occupancy, when reviewing a kernel change for resource impact, or as a fast pre-check before spending a profiling run. Usage: /isa-resource-diff [<test-or-command>] [--arch <gfx>]
    0
    installs
36. ▌
    Add Target Atom Op · rocm

    Add a new target-specific Mma / Copy Op type to a FlyDSL backend dialect (`lib/Dialect/Fly<TARGET>/<SUBTARGET>/` + `include/flydsl/Dialect/Fly<TARGET>/IR/`). Covers the MmaOp/CopyOp type design, the stateful-vs-stateless variants, and the `emitAtomCall` / `emitAtomCallSSA` lowering contract to the backend dialect (LLVM/ROCDL/NVVM/SPIR-V/...). Use when adding a new tensor-core / matrix instruction (MFMA, WMMA, HMMA, WGMMA, ...), a new buffer / shared-memory / global copy atom, a new stateful copy (per-atom offset or descriptor), or bringing up a new backend dialect (`FlyPTX`, `FlyCPU`, ...). Reference implementations live in `FlyROCDL`. Usage: /add-target-atom-op
    0
    installs
37. ▌
    Prefetch Data Load · rocm

    Apply prefetch optimization to FlyDSL kernel loops: pre-load the first iteration's data before the loop, issue async loads for the next iteration inside the loop body, and swap buffers at the loop tail via runtime loop-carried values. This overlaps data load latency with compute instructions. Use when a kernel has a loop where buffer_load feeds into MFMA/compute and load latency is exposed. Usage: /prefetch-data-load
    0
    installs
38. ▌
    Debug Flydsl Kernel · rocm

    Debug FlyDSL GPU kernels that produce NaN, inf, wrong results, or crash. Covers cache invalidation, tracing pitfalls (runtime conditionals, range vs range_constexpr), loop-carried state packing, buffer_load addressing, MFMA operand layout verification, LDS bank conflict diagnosis, and systematic error isolation (all-1s test, single-partition test, host-side tensor inspection). Use when a FlyDSL kernel produces incorrect output or compilation errors. Usage: /debug-flydsl-kernel
    0
    installs
39. ▌
    Kernel Code Cleanup · rocm

    Clean up FlyDSL kernels and shared helpers while preserving numerical behavior and performance. Use for raw-IR migrations, helper deduplication, dead-code removal, and reviews of those changes against the checkout API.
    0
    installs
40. ▌
    Capture Kernel Trace · rocm

    Capture GPU kernel ATT (Advanced Thread Trace) via rocprofv3 on a remote Docker container or locally. Discovers kernel names, configures input.yaml with the target kernel_include_regex, runs rocprofv3 -i input.yaml with FLYDSL_DEBUG_ENABLE_DEBUG_INFO=1, and downloads the latest ui_output_agent_* directory for analysis. Usage: /capture-kernel-trace <test_script.py> [kernel_name_pattern]
    0
    installs
41. ▌
    Kernel Trace Analysis · rocm bundle

    Profile GPU kernels using rocprofv3 to collect ATT instruction-level traces, then analyze the trace data using hotspot_analyzer.py to identify top-K stall hotspots (VMEM-load, VMEM-wait, LDS/SMEM-wait, barrier, MFMA stalls) mapped back to source lines, and produce an actionable optimization plan. Usage: /kernel-trace-analysis <cmd> Can also analyze an existing dispatch dir directly: /kernel-trace-analysis --dir <path>
    0
    installs
42. ▌
    Bisect Perf Regression · rocm

    Find the exact commit that caused a GPU kernel performance regression using binary search (git bisect). Given a good commit (fast), a bad commit (slow, defaults to HEAD), and a benchmark command, automatically checks out commits, runs the benchmark, extracts the metric, and narrows down to the offending commit. Reports the regression commit with its diff and suggested root cause. Usage: /bisect-perf-regression <good_commit> [bad_commit] -- <bench_cmd>
    0
    installs
43. ▌
    Flydsl Kernel Authoring · rocm

    Comprehensive reference for authoring FlyDSL GPU kernels on AMD GPUs. Covers the layout algebra, tiled copy/MMA, buffer ops, loop-carried range loops, SharedAllocator (LDS), autotuning, and common patterns. Use when writing, reviewing, or understanding FlyDSL kernel code.
    0
    installs
44. ▌
    Flydsl Tile Programming · rocm

    Guided step-by-step wizard for producing a new FlyDSL GPU kernel from a requirement: classify the kernel type, pick a skeleton, fill in compute, add control flow / sync / LDS, then test on GPU. Use when the user wants to WRITE a new kernel, port a Triton kernel to FlyDSL, or learn tile programming by following a procedure. For API/layout-algebra lookups, per-op reference tables, and troubleshooting, use the flydsl-kernel-authoring skill instead.
    0
    installs
45. ▌
    Decode And Triage · rocm bundle

    Decode and triage HRR capture archives with full GPU replay by default (Linux). Builds hrr-playback when missing. On Windows: full native GPU replay via triage_archive.ps1 + ensure_playback.ps1 (PowerShell); Docker replay requires Linux or WSL2. Never edits source. Print finding summary in the chat reply.
    0
    installs
46. ▌
    Validate · rocm

    Run full rocDecode build and test pipeline after code changes
    0
    installs
47. ▌
    Pc Sampling · rocm bundle

    Profiles a target application using rocprofv3 with program counter sampling enabled and then analyzes the results. Use when the user has a program that runs on an AMD GPU and asks to perform PC sampling, to determine the runtime performance characteristics of their application, to determine stall reasons, or to determine hotspots in the code.
    0
    installs
48. ▌
    Writing Plans · rocm

    Use when an approved spec exists and you need a bite-sized, file-level implementation plan before any code is written. Produces a plan ready for executing-plans or subagent dispatch.
    0
    installs
49. ▌
    Perf Check · rocm

    Run the rocDecode performance test and compare to baseline to catch regressions
    0
    installs
50. ▌
    Amdsmi CI Logs · rocm bundle

    Use when a GitHub Actions run or job for an amd-smi / rocm-systems PR failed and the web UI shows truncated output, when you need the full failing log, or when you need CI artifacts (test reports, packages). Triggers: an actions/runs/<id>/job/<id> URL, 'where are the logs', 'output is truncated', 'where are the artifacts', a red check on a PR.
    0
    installs
51. ▌
    Writing Skills · rocm

    Use when creating or editing a SKILL.md file under .claude/skills/. Defines the TDD-for-documentation discipline, structure, and CSO rules that every amd-smi skill must follow.
    0
    installs
52. ▌
    Rdc Build Install · rocm

    Build and install RDC from source. Use when: building locally, installing before tests, pre-review build verification, build + install + verify. Requires GRPC_ROOT to be set.
    0
    installs
53. ▌
    Executing Plans · rocm

    Use when a written implementation plan exists at ${TMPDIR:-/tmp}/amdsmi-agent-plans/ and you need to execute it task-by-task in the current session with verification at each step.
    0
    installs
54. ▌
    Cuid Build Install · rocm

    Build and install cuid from source. Use when: building locally, installing before tests, pre-review build verification, build + install + verify.
    0
    installs
55. ▌
    Amdsmi Interrogate · rocm

    Use whenever amd-smi work begins from a design — a Confluence Feature Design, a Jira/SWDEV ticket, a driver hand-off, or a user description. Reconciles the design against what the code actually does, questions it adversarially, and produces an approved spec. This is the single front door to any feature or behavior change.
    0
    installs
56. ▌
    Amdsmi Test Runner · rocm bundle

    Run C++ and Python tests for amd-smi. Use when: running tests, verifying test results, checking test coverage, pre-review test validation.
    0
    installs
57. ▌
    Populate Project Map · rocm

    Creates or updates a human-readable project map for the cuid workspace. Use when onboarding, when workspace layout changes, or when agents need up-to-date layout info.
    0
    installs
58. ▌
    Feature Unit Testing · rocm bundle

    Use when writing, planning, or improving unit tests for low-level transport or systems code — especially when reasoning about branch coverage, test gaps, identifying which uncovered paths are worth pursuing, or deciding when a feature's test suite is ready to merge.
    0
    installs
59. ▌
    Amdsmi Agent Handoff · rocm

    Use when handing work from one amd-smi agent to another (planning→development, planning→review, development→review) or compacting a long session into a fresh one. Produces a compact handoff doc referencing artifacts by path instead of duplicating them.
    0
    installs
60. ▌
    Amdsmi Build Install · rocm

    Build and install amd-smi from source. Use when: building locally, installing before tests, pre-review build verification, build + install + verify.
    0
    installs
61. ▌
    Systematic Debugging · rocm

    Use when encountering any bug, test failure, build failure, or unexpected behavior in amd-smi — before proposing any fix. Enforces root-cause investigation before symptom patching.
    0
    installs
62. ▌
    Amdsmi Packaging Test · rocm

    Integration testing for amd-smi Python packaging. Use when: testing package installs, multi-OS testing, manylinux wheel builds, offline install verification, glibc/Python version compatibility, post-install smoke tests.
    0
    installs
63. ▌
    Pressure Testing Skills · rocm

    Use when validating or hardening a SKILL.md, prompt, rule, or any agent-followed document — when you need proof it actually works, not just that it reads well. Runs a fresh subagent that follows the doc literally on a known-answer fixture, captures friction, and iterates to determinism before minimizing. Use after writing-skills produces a draft, or whenever a skill 'looks right' but hasn't been proven under an agent.
    0
    installs
64. ▌
    Test Driven Development · rocm

    Use when implementing any amd-smi feature, bug fix, or behavior change — before writing implementation code. Enforces strict RED-GREEN-REFACTOR: failing test first, watch it fail, minimal code to pass, refactor.
    0
    installs
65. ▌
    Amdsmi Python Style Guide · rocm

    ROCm Python style guide based on TheRock. Use when: writing Python code, reviewing Python PRs, checking Python style, creating Python scripts, type hints, error handling patterns, pathlib usage, argparse CLI design.
    0
    installs
66. ▌
    Amdsmi Output Verification · rocm

    A checklist for verifying whether AMD SMI is reporting GPU information correctly.
    0
    installs
67. ▌
    Amdsmi Restructure Commits · rocm

    Use when finishing an amd-smi development branch — consolidating commits into logical groups with clean messages AND deciding how to integrate the work (merge to develop, push and open PR, keep as-is, or discard). Covers commit restructuring plus the merge/PR/cleanup workflow.
    0
    installs
68. ▌
    Amdsmi Using Git Worktrees · rocm bundle

    Use when starting amd-smi feature work, executing an implementation plan, or reviewing a PR that needs an isolated workspace away from the main checkout. Sets up a worktree following the rocm-systems-pr<PR#> convention.
    0
    installs
69. ▌
    Amdsmi Changelog Automation · rocm

    Check and generate changelog entries for amd-smi. Use when: reviewing PRs for changelog updates, generating release notes, checking CHANGELOG.md compliance.
    0
    installs
70. ▌
    Dispatching Parallel Agents · rocm

    Use when facing 2+ independent problems with no shared state — e.g., unrelated test failures in different subsystems, multiple independent bug investigations, parallel research tasks. Dispatch one focused subagent per domain instead of investigating sequentially.
    0
    installs
71. ▌
    Amdsmi Improvement Evaluation · rocm

    A structured approach to evaluating AMD SMI's current state and identifying improvement opportunities.
    0
    installs
72. ▌
    Verification Before Completion · rocm

    Verification checklist before marking work complete. Use when: finishing a fix, completing a review finding, validating a build, confirming a refactor.
    0
    installs
73. ▌
    Amdsmi Commit And Pr Conventions · rocm

    Use when writing or restructuring git commits or opening/updating a pull request for amd-smi — composing commit titles, commit message bodies, PR titles, or PR descriptions. Defines the Conventional Commits `type(amdsmi):` title convention enforced by the Systems PR Bot, the rocm-systems PR template sections, the unit-test and JIRA/ISSUE-reference gates, brevity caps, and the rule that JIRA tickets appear only in the PR JIRA ID section, never in code comments or commit bodies.
    0
    installs
74. ▌
    Ask · rocm

    Use when user asks a question that doesn't require any action - just explanation or information
    0
    installs
75. ▌
    Git · rocm

    Git workflow dispatcher - routes to specialized git skills for commits, PRs, and reviews
    0
    installs
76. ▌
    Watch · rocm

    Schedule a recurring poll of an external condition (PR merge, CI green, Jira state, background build) that runs a follow-up action when met and self-stops. Use for async waits across sessions.
    0
    installs
77. ▌
    Notify Me · rocm

    Send a push notification to the user's phone via ntfy.sh when a condition is met, by composing a watch cron whose action is a curl POST.
    0
    installs
78. ▌
    Pr Review · rocm bundle

    Reviews Pull Requests or local diffs with an 8-agent fan-out covering static analysis, dead code, code smells + quality (naming, complexity, single-responsibility, magic numbers), language rules (C++/Python/CMake), architecture, simplification, performance (hot-path classification, allocations, locks, I/O), and undefined behaviour (signed overflow, lifetime, strict aliasing, data races, sanitizer coverage; C/C++/unsafe-Rust only). Use when the user asks to "review this PR", "review the diff", "audit this branch", "/pr-review", or when staging changes before push.
    0
    installs
79. ▌
    Git Commit · rocm

    Create meaningful git commits with well-structured messages - analyzes changes, generates descriptive subject and body, follows conventional commits
    0
    installs
80. ▌
    Rocprofsys · rocm

    ROCm Systems Profiler (rocprofiler-systems) project workflows - configure, build, test, and development
    0
    installs
81. ▌
    Code Smells · rocm

    Detect code smells based on refactoring.guru catalog - use when analyzing code for anti-patterns and refactoring opportunities
    0
    installs
82. ▌
    Radisha Help · rocm

    Full workflow reference and detailed rules - invoke when you need comprehensive guidance on how to use radisha skills
    0
    installs
83. ▌
    Git Gh Client · rocm

    Core GitHub CLI operations - check availability, search PRs, get PR details, check status. Invoked by other git skills.
    0
    installs
84. ▌
    Planning Base · rocm

    Base planning skill with shared rules - do not invoke directly, use specific planning skills instead
    0
    installs
85. ▌
    Planning Docs · rocm

    Planning skill for documentation - no changelog updates
    0
    installs
86. ▌
    Radisha Skills · rocm

    List all available radisha skills with their command aliases, organized by category
    0
    installs
87. ▌
    Radisha Update · rocm

    Update radisha skills to the latest version - works with symlinks, git clones, and submodules
    0
    installs
88. ▌
    Testing Pytest · rocm

    Pytest testing skill with modern patterns - use for writing Python tests with fixtures, parametrization, and modular design
    0
    installs
89. ▌
    Library Amd Smi · rocm

    AMD SMI C++ library for GPU/CPU/NIC monitoring and management. Use when working with AMD hardware monitoring, GPU temperature, power, memory, clocks, PCIe, XGMI, SDMA (System DMA), AINIC (AI NIC) network interfaces, or any amdsmi.h functions.
    0
    installs
90. ▌
    Planning Bugfix · rocm

    Planning skill for bug fixes - includes optional changelog update
    0
    installs
91. ▌
    Programming Cpp · rocm

    C++ programming skill based on C++ Core Guidelines - use for implementing C++ code
    0
    installs
92. ▌
    Static Analysis · rocm

    Run static analysis tools on changed files and generate a report
    0
    installs
93. ▌
    Planning Feature · rocm

    Planning skill for new features - includes changelog summary and test case consideration
    0
    installs
94. ▌
    Rocprofsys Build · rocm

    Build, test, and install rocprofiler-systems after configuration
    0
    installs
95. ▌
    Testing Testplan · rocm

    Create test plan files for developer verification and QA handoff - use after implementation is complete
    0
    installs
96. ▌
    Planning Refactor · rocm

    Planning skill for refactoring - improves existing code following best practices, reduces duplication, improves readability and testability
    0
    installs
97. ▌
    Programming Python · rocm

    Python programming skill based on PEP 8 and modern Python best practices - use for implementing Python code
    0
    installs
98. ▌
    Verify Pmc Metrics · rocm

    Mock-based rigorous PMC metric validation — tests sentinel filtering, value ranges, and architecture-specific behavior across Perfetto and RocPD
    0
    installs
99. ▌
    Analyze Pr Comments · rocm

    Use when user wants to list, analyze, review, or summarize GitHub PR comments on a pull request number or URL
    0
    installs
100. ▌
     Review Architecture · rocm

     Use when user wants architecture documentation covering component interactions, hierarchy, expansion points, or system weaknesses - for single components, integrations, or whole systems
     0
     installs