
Performance Profile
Finds performance bottlenecks in a game codebase and ranks fixes against set budgets.
- Repository
- Donchitos/Claude-Code-Game-Studios
- Stars
- 32,965
- Licence
- MIT
- Files
- 1 file
What it does
When installed, the agent checks whether there is enough real data (profiler output, budgets, or code to analyze) before reporting anything. If the data is missing, it says so instead of inventing numbers. When data exists, it looks through the code for likely slow spots such as per-frame update functions, memory allocation patterns, rendering costs, and loading/network I/O, then writes a ranked list of optimization recommendations to a report file, asking permission first.
Great for
- Game developers who want a structured review of frame-time, memory, or draw-call risks against their own budgets
- Teams using the Claude Code Game Studios setup who want a report saved under production/polish for comparison across runs
Skip it if
- Projects with no performance budgets or profiler data: the skill is designed to return NOT ASSESSED rather than a pass, so there is little useful output without them
- Non-game software or anyone needing live profiling: it reads files and code, it does not run a real profiler itself, so its estimates are inferences rather than measurements
Try asking
- “/perf-profile full”
- “/perf-profile combat-system”
- “Profile the inventory system against our frame budget and rank the fixes”
- “Run a full performance review and save the report for the Polish gate”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
perf-profile.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip perf-profile.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip perf-profile.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Set performance.target_framerate, frame_budget_ms, draw_call_limit and memory_ceiling_mb in project.yaml so budgets are read from committed config rather than prose
- Choose the enforcement mode carefully: block makes over-budget systems release-stopping and can fail the /gate-check Polish gate
- Name a specific system for a focused run, or use full for a broader sweep
What's inside
- SKILL.md
Redistributed unmodified from Donchitos/Claude-Code-Game-Studios under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.