add-op
OfficialStreamline adding a new op to tt-mlir.
Authortenstorrent
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill provides a comprehensive, repeatable blueprint for adding a new operation (op) to the tt-mlir compiler stack, covering TTNN dialects, TTIR dialects, conversions, EmitC/EmitPy, flatbuffers, runtime, and tests to ensure end-to-end integration.
Core Features & Use Cases
- Structured, end-to-end op addition workflow spanning TTNN definitions, TTIR mappings, TTIR→TTNN conversions, TTNN→EmitC/EmitPy, flatbuffer schema, runtime wiring, and test scaffolding.
- Guidance anchored to the TTNN type mappings in references/ttnn_type_mapping.md to align C++ API signatures with MLIR definitions.
- Comprehensive testing scaffolds including TTIR-to-TTNN conversion, StableHLO composites (where applicable), and EmitC/EmitPy pipelines to validate the op across toolchains.
Quick Start
Define the new op in TTNN, propagate it through TTIR and all conversions, then implement runtime and tests as described.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: add-op Download link: https://github.com/tenstorrent/tt-mlir/archive/main.zip#add-op Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.