Dogacel

Dogacel/auto-gpu-kernel

Agent Claude Code

Benchmark the current Triton kernel and identify compute/memory/launch bottlenecks with concrete numbers.

Dogacel/auto-gpu-kernel

Agent Claude Code

Clean-context diagnosis for planning next experiments.

Dogacel/auto-gpu-kernel

Agent Claude Code

Inspect workload input characteristics (shapes, distributions, padding, locality) and recommend kernel optimizations keyed to them.

Dogacel/auto-gpu-kernel

Command Claude Code

Pipe output to a file (e.g. bench.log in repo root) — /log-experiment will attach it.

Dogacel/auto-gpu-kernel

Command Claude Code

Log the most recent experiment. Never skip — failures are as valuable as wins.

Dogacel/auto-gpu-kernel

Command Claude Code

Iteratively improve solution/triton/sparsefused.py. Rules in CLAUDE.md are non-negotiable.

Dogacel/auto-gpu-kernel

Agent Claude Code

Characterize the workload inputs — shapes, distributions, padding, locality — and recommend optimizations keyed to them. Call when you suspect a data-shape angle.

Dogacel/auto-gpu-kernel

Skill Claude CodeCodex

Record the most recent optimization attempt — snapshot the kernel, write result.md, append to the summary index. Use after every benchmark, including failures.

Dogacel/auto-gpu-kernel

Skill Claude CodeCodex

One autonomous kernel optimization iteration — plan a single change, implement, benchmark, log. Use for the main optimization loop.

Dogacel/auto-gpu-kernel

Skill Claude CodeCodex

Generate the repository-specific validation and benchmark adapters that kbench will orchestrate. Use once when harness/prepared.json is absent.

Dogacel/agent-blog

Plugin Claude Code

Bundles 4 commands, 2 hooks, 1 MCP server · 70 tokens together

Your AI agent automatically writes and publishes blog posts about things it finds interesting during coding sessions.