OpenSkillIndex← back to search
claude-codedevops

Claude-skill-registry fusion-perf

Compare performance optimization options and choose the most effective; use when tuning or speeding up a system.

install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/fusion-perf" ~/.claude/skills/majiayu000-claude-skill-registry-fusion-perf && rm -rf "$T"
manifest: skills/data/fusion-perf/SKILL.md
tags
#performance-tuning#benchmarking#parallel-processing#system-optimization
source content

Fusion Performance

Overview

Use an F-thread: three workers implement different optimizations in separate worktrees. The queen compares results.

Inputs

  • Performance problem and metrics

Workflow

  1. Verify 
    git
    and 
    mprocs
    .
  2. Create session variables and worktrees.
  3. Write 
    tasks.json
    , worker prompts, and queen prompt.
  4. Launch mprocs.

Worktree Commands

git worktree add "{WORKTREE_ROOT}/impl-a" -b fusion/{SESSION_ID}/impl-a
git worktree add "{WORKTREE_ROOT}/impl-b" -b fusion/{SESSION_ID}/impl-b
git worktree add "{WORKTREE_ROOT}/impl-c" -b fusion/{SESSION_ID}/impl-c

Worker Prompt Outline

  • Worker A: algorithmic improvements
  • Worker B: caching or batching
  • Worker C: concurrency or IO tuning

Queen Prompt Outline

  • Compare benchmarks, complexity, and regression risk

mprocs Launch

mprocs --config .hive/mprocs.yaml

Output

  • Selected optimization strategy with benchmarks