SuperAPI AI DirectoryFind an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-15 00:13
41,133 results
Can complete the installation and deployment of Ascend NPU drivers and firmware, implementing regular expression matching and extraction of installation packages, on-demand addition of executable permissions, dual Python+Shell package verification, prior verification before installation of system dependencies, and adaptation to CentOS/RHEL/Ubuntu/Debian systems. Suitable for Ascend NPU driver and firmware installation and deployment.
Ascend C code review skill. Conduct security specification reviews of code based on hypothesis testing methodology. When invoking, you must explicitly provide: code snippet and review rule description. TRIGGER when: the user requests code review, code review, asks about code security issues, checks coding standards, or needs to check specific code issues (such as memory leaks, integer overflow, null pointers, etc.). Keywords: Ascend…
Compile and install the AscendC operator and run accuracy tests. TRIGGER when: after operator code generation is complete and compilation verification, whl package installation, accuracy test execution is needed, or when compilation/testing fails and troubleshooting is required. Keywords: build.sh, compilation, installation, whl, pytest, accuracy test, compilation error, NPU test.
AscendC operator end-to-end development orchestrator. Use when the user needs to develop a new operator, implement a custom operator, or complete the full process from requirements to testing. Keywords: operator development, operator development, end-to-end, complete process, workflow orchestration, new operator.
Generate PyTorch-style interface documentation (README.md) for AscendC operators. Trigger scenarios: use when interface documentation needs to be generated after compilation and debugging pass, or when the user mentions "generate operator documentation", "create README", "document operator", "help me write documentation" (operator context), or "operator documentation".
AscendC operator precision issue debugging and root cause localization. Use when operator precision tests fail (allclose fails, result deviation, output all zeros/NaN, etc.). Process: error distribution analysis → code error-prone point review → experimental isolation → printf/DumpTensor instrumentation → fix verification. Keywords: precision debugging, precision issue, inconsistent results, error localization, allclose…
Initialize the AscendC operator project and create a compilable operator skeleton. Trigger scenarios: (1) the user requests creating a new operator; (2) keywords: ascendc operator, create operator, operator directory, operator initialization; (3) need to quickly implement based on the ascend-kernel template. This skill does not only create directories, but also outputs standard files and a checklist that are "ready for continued development".
Generate operator project deliverables based on the CATLASS operator design document
Analyze official Megatron-LM commits, PRs, and branch change sets to identify feature evolution, candidate breaking changes, and migration-relevant events. Use…
Track and normalize change requests against the official Megatron-LM repository by branch, PR, commit, commit range, or time window. Use when Codex needs to…
Python code refactoring skill, covering code smell identification, design pattern application, readability improvement, and practical experience. Use this skill when the user requests "refactor code", "refactor", "code optimization", "improve code quality", "code smell review", "apply design patterns", "improve readability", or submits a code review request. Supports outputting a structured refactoring document after refactoring is complete ("output refactoring document"…
Generate Triton kernel code based on an Ascend NPU operator design document (or direct requirements). Use when the user needs to implement a Triton operator or convert a design document into executable code. Core outputs: kernel code + basic correctness test. Keywords: Triton kernel, operator implementation, code generation, code generation.
Generate Triton operator requirement documents for Ascend NPU. Use when users need to design new Triton operators, write operator requirement documents, or perform operator performance optimization design. Core outputs: function definition, API interface, tiling strategy, kernel implementation plan.
Ascend Triton operator full-process development orchestration. Use when the user needs to develop a Triton operator from scratch, carry out an end-to-end development process, or is unsure which sub-skill to use. Automatic orchestration: environment configuration→requirements design→code generation→static review→precision verification→performance evaluation→performance optimization. Keywords: full process, development orchestration, end-to-end, workflow orchestration.
Generate standardized interface documentation for Ascend NPU Triton operators. Use when users need to create a README, generate API documentation, write product support tables, or organize parameter descriptions for operators. Keywords: documentation generation, doc generation, README, interface documentation, API documentation.
Verify and build the environment required for Triton operator development on the Ascend platform, including CANN, Python/torch/torch_npu/triton-ascend dependencies, and PATH environment variable settings. Use when the user needs to configure the Triton operator development environment, check the CANN/torch/triton-ascend installation, or verify whether the environment is usable.
Review code for correctness, maintainability, and security risks. Use when preparing a PR or auditing a change set.
Sub-skill — custom editing interface. Use when the user wants to make a small "operate then export" tool in HTML — drag-and-drop triage board, feature flag editor (outputs a diff), prompt tuner (live preview, outputs a prompt). Every editor **must** have a "copy result" button so the UI…
Use when the user wants to wrap up the current coding session, summarize what happened in this session before ending work, document learnings and decisions,…
Compiles iOS Xcode projects via xcodebuild. Use when building iOS apps, running xcodebuild, compiling Xcode projects, CI/CD for iOS, or when the user mentions…
Build professional 3D maps and GIS applications with MapV-Three - a Three.js-based 3D map rendering engine that supports a complete set of map development capabilities including sky environments, base map loading, 3DTiles, visualization components, LBS services, model assets, GIS analysis, editing and measurement, and more. Suitable for creating Web applications such as 3D digital twins, urban visualization, and GIS analysis.
Answer questions about the GM Farcaster podcast archive (Farcaster, Base, and the wider crypto-social ecosystem) using the paid Warpee Knowledge API. Use this…
Launch, buy, sell, and claim fees for hood.markets tokens on Robinhood Chain (4663) via api.hood.markets. Use for hoodmarkets, hood.markets, $hood, launch…
Structured frameworks for documenting architectural decisions using MADR templates, option analysis, trade-off matrices, and human-in-the-loop AI assistance.