Loading...
Loading...
Found 2,732 Skills
Catlass Operator End-to-End Development Orchestrator. Based on ascend-kernel (csrc/ops), it connects catlass design, catlass-operator-code-gen and ascendc sub-skills to complete the closed loop from project initialization to documentation, precision, and performance. Keywords: Catlass, end-to-end, ascend-kernel, operator development, workflow orchestration.
Verify and build the required environment for Triton operator development on the Ascend platform, including configurations of dependencies such as CANN, Python/torch/torch_npu/triton-ascend and PATH environment variables. This is used when users need to configure the Triton operator development environment, check the installation of CANN/torch/triton-ascend, or verify whether the environment is available.
Accepts Triton operator implementations, automatically invokes Torch small operator implementations (CPU or NPU) for precision comparison, and generates precision reports. It is used when users need to verify the correctness and precision of Triton operator implementations, compare precision with PyTorch implementations, and generate standardized precision reports.
Validate, format, and convert between JSON, YAML, and TOML. Parse and query structured data files. No API key required.
Format and validate code in various languages. Python, JavaScript, JSON, YAML, Markdown, and more. Uses standard formatters when available.
Schedule and manage cron jobs. Use when: user needs to create, list, remove, or test scheduled tasks.
Execute read-only SQL queries against Databricks. Use when you need to run a specific SQL query, aggregate data, join tables, or answer analytical questions about Databricks data.
guide and workflow for rule-creation.
Runs the Metabase semantic checker against a tree of Representation Format YAML files to verify that all references resolve — cross-entity references (collection_id, dashboard_id, parent_id, parameter source cards, snippet references, transform tags, etc.) and references to columns inside MBQL and native queries. Use when the user asks to "semantic check", "check references", "validate queries against the schema", or diagnose a broken reference. Requires database metadata on disk (by default `.metabase/metadata.json`).
Zero-context verification that every number, comparison, and scope claim in the paper matches raw result files. Uses a fresh cross-model reviewer with NO prior context to prevent confirmation bias. Use when user says "审查论文数据", "check paper claims", "verify numbers", "论文数字核对", or before submission to ensure paper-to-evidence fidelity.
Background knowledge for droid-control workflows -- not invoked directly. Deliverable verification against commitments.
Use when creating or editing any prompt (commands, hooks, skills, subagent instructions) to verify it produces desired behavior - applies RED-GREEN-REFACTOR cycle to prompt engineering using subagents for isolated testing