在 Ascend 950 / DAV_3510 上用户明确指定 Tensor API 时的实现路线权威源（硬约束，不由 Blaze 优先选型改写），Tensor API 为主、C API 为辅，覆盖 API 分工、Layout、内存预算、工具链、精度与性能。
109 skills
16
claudecodex
catlass-dsl-generator
基于 Markdown 任务契约的精简 CATLASS DSL 算子开发优化工作流。
Not in a marketplace yet, so it loads for one session with --plugin-dir.
Plugin catlass-dsl-generator in hicann/cannbot-skills. Commands read from the repo's manifest — today.
What it installs
Skills (7)
catlass-dsl-benchUse when independently checking and timing a CATLASS Python DSL solution against definition and workload files; this version provides the measurement contract and requires the agent to write and run its own benchmark script.
catlass-dsl-designUse when specifying a new CATLASS Python DSL operator and producing its Torch oracle, definition, workloads, executable reference evidence, and frozen Markdown preliminary design
catlass-dsl-developGenerate a complete CATLASS Python DSL operator from a frozen DESIGN.md, then debug, review, test, benchmark, profile, and optimize it with knowledge queries and a text-tracked milestone log. Use for CATLASS DSL operator implementation, debugging, and delivery from a frozen design.
catlass-dsl-knowledge-upgrade维护 catlass-dsl-generator 知识库与版本基线。当用户提到 catlass 新版本发布、v2.3、ascend-catlass-dsl 升级、知识库更新、API 矩阵、样例路径失效或知识库漂移时使用；本版本不使用状态机脚本，按文本清单推进。
catlass-dsl-knowledgeUse when querying, validating, or appending evidence-backed CATLASS DSL kernel knowledge for Ascend NPU operator development in an OKF v0.2 project bundle; this version reads and writes Markdown files directly without scripts.
catlass-dsl-optimizeIteratively optimize, speed up, benchmark, and profile one existing CATLASS DSL operator source file on Ascend NPU with a text-tracked round ledger, noise-aware best selection, and profiler-guided tuning. Use for CATLASS DSL performance optimization or repeated kernel tuning.
catlass-dsl-workflowUse when the user asks for any CATLASS DSL operator task (design, implementation, debugging, benchmark, optimization, knowledge) or mentions catlass.tla, to route stages and apply the gate-checked workflow rules before any other catlass-dsl skill runs. 当用户提及 CATLASS DSL 算子开发、实现、复现、调试、精度对齐、benchmark、
Problems found in this repo's manifests
- "tilelang-op-orchestrator" points to ./plugins-official/tilelang-op-orchestrator, which doesn't exist in the repo — install will fail
- "tilelang2ascendc-ops-generator" points to ./plugins-community/tilelang2ascendc-ops-generator, which doesn't exist in the repo — install will fail
More from hicann/cannbot-skills
All 29 plugins →ops-code-reviewer
claudeAscend C 代码检视 Skill。支持文件检视、PR 检视和快速定向排查。
109 skills
16
ops-direct-invoke-skills
claudeAscend C Kernel 直调算子开发 Skills（40+ 个）。API 检索、Tiling 设计、Tensor API/Blaze 路线、精度调试、性能采集、环境检查、代码模板等。
109 skills
16
Ascend C 算子性能进化优化共享 Skills（2 个）。NPU 架构与芯片规格查询、性能采集与分析。
109 skills
16
ops-perf-optimize-skills
claudeAscendC 算子性能调优 Skills（6 个）。性能仿真、Profiling 采集、Tiling 建模与流水分析、性能优化最佳实践、MC² 最佳实践、API 文档检索。
109 skills
16
Ascend C 注册调用算子开发 Skills（21 个）。API 检索、Tiling 设计、精度调试、性能采集、环境检查、代码模板、测试设计、UT 开发等。
109 skills
16
Similar plugins
mindspeed-llm-skills
claudeMindSpeed-LLM skill set for Huawei Ascend NPU, covering the full training pipeline: environment setup, data preprocessing, weight conversion, distributed training, and end-to-end orchestration. Use when deploying MindSpeed-LLM training on Ascend NPU.
191 skills
ascend-ai-coding/awesome-ascend-skills 184
vllm-omni-hardware
claudeHardware backend guidance for CUDA, ROCm, NPU, and XPU.
hsliuustc0106/vllm-omni-skills 89
ring
claudeControl-plane plugin for tracking rings: recursive units of work, each a problem to settle rather than a session. A ring has a Brief frozen at open, a Trace of local working memory discarded at close, and a Delta typed by effect (ESTABLISHED, INVALIDATED, SURFACED, DELIVERABLE) that crosses exactly one turn to its parent. Ring state is orthogonal to what carries it: the four vehicles are inline, subagent, session and resumed. For vehicle session, ring layers on top of the team plugin as its vehicle — team keeps circulating mandates, reports and relay packets between named agents unchanged, while ring tracks which problem is open, who is carrying it, and what crosses back up. Ships four skills (open, close, rings, sync) for descending into and ascending out of a ring, listing/graphing the live tree, and persisting a ring's state so it survives a /clear, an on-disk writer token (carrier:) taken over by resuming a parked or active ring, exclusive create:/writes: write-right tracking with prefix-inclusion collision detection and a lineage exception for parent-child delegation, and a closing sweep that scans disk for files born or touched outside declared claims.
4 skills
digital-stoic-org/agent-skills 21
npu-skillyard
claudePyTorch algorithm -> Ascend PTO-ISA kernel generation pipeline (decompose, generate, validate, benchmark, optimize, fuse).
6 skills · 3 agents · 1 MCP server
huawei-csl/npu-skillyard 8
ascend-migration
claudePyTorch to Ascend NPU migration analyzer and optimizer. Analyze models for Ascend compatibility and generate fully optimized migrated code with ALL performance optimizations (fused attention, torch.compile, data transfer optimization, etc.).
6 skills · 3 commands
FeRhodium/ascend-migration 7
CPU, GPU, NPU and Memory Microarchitecture: architecture versus microarchitecture, pipelining, out-of-order execution, branch prediction, execution units and SIMD, cache organization, cache coherence, memory consistency models, virtual memory and translation, prefetching, GPU microarchitecture and memory systems, NPUs and dataflow architectures, numeric formats, DRAM internals, memory controllers, emerging memory interfaces, power and clocking, microarchitectural security, ISA design, simulation and modelling, measurement, roofline and fundamental limits, and specialization.
6 skills
the-vibey-project/vibey 5
Maintain catlass-dsl-generator?
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.
claude & codex plugins29 plugins