Plan Performance — Development skill for Claude Code
Measure a performance problem and plan verified fixes.
How to install Plan Performance
Installs to ~/.claude/commands/polewskm-project-template-plan-performance.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/polewskm/project-template/HEAD/.claude/commands/plan-performance.md -o ~/.claude/commands/polewskm-project-template-plan-performance.md Restart Claude Code, or start a new session, for it to be picked up.
What Plan Performance does
description: "Measure a performance problem and plan verified fixes" argument-hint: "[What is slow, how slow, and the target]" model: opus effort: high disable-model-invocation: true
Plan in a worktree of its own
Each plan lives in its own git worktree, so several Claude Code sessions can plan and execute different plans on this machine at once. Executing, re-planning and reviewing this plan later happen in the same worktree. Set it up before you read any code, so your research match
Alternatives in Development
- Skill Creator — Create new skills, modify existing skills, and measure skill performance 94.1k ★
- Agent.System.Main.Solving — Problem solving 19k ★
- Create PRD — Create a PRD with 8-section template covering problem to release 7.8k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
FireFinder
A shared memory of verified fixes for Claude. When someone confirms a fix worked, FireFinder keeps the general
Maintain Nt
Maintenance sweep — stale deps / Actions / advisories / dead links / CWV drift, ranked into a fix-workplan; ap
Bobby Performance
Performance benchmarking — measure and detect regressions
Perfetto Android Skills
Agent skill: measure, diagnose, fix and re-measure Android app performance from a Perfetto trace
Pbi Explain
Deconstruct any DAX measure into a structured plain-English breakdown covering business logic, filter context
Pbi Optimise
Run a 14-rule performance scan on any DAX measure, identifying anti-patterns like unnecessary iterators, FILTE
Related Agents
Triz Effects
Effects-and-information-fund specialist for the /triz workflow. Use it when the problem reduces to "we need a
Performance Monitor
Monitor agent execution, track token usage, measure response quality, and optimize workflows
QA Performance Tester
Performance test engineer. Use to establish baselines and budgets, write and run load/stress/spike/soak tests