aryandaga7

Autoperf AI — AI skill for Claude Code

AI community

Autonomous AI agent performance optimizer — eval, optimize, and measure your AI SDK agents.

How to install Autoperf AI

This entry records only its repository, not the path inside it, so there is no exact command to give. Open aryandaga7/autoperf-ai and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Autoperf AI does

Autonomous AI agent performance optimizer — eval, optimize, and measure your AI SDK agents.

Alternatives in AI

Deepreasoning — A high-performance LLM inference API and Chat UI that integrates DeepSeek R1's CoT reasoning traces with Anthr 5.4k ★
Vllm Mlx — High-performance OpenAI and Anthropic compatible LLM inference server for Apple Silicon 1.5k ★
Late CLI — High-performance AI agent for long-horizon tasks and local models 442 ★

README

Autonomous agent optimization, powered by Claude Code

AutoPerf evaluates your AI agent's cost, quality, and latency, then spawns autonomous Claude Code agents in isolated git worktrees that research optimization strategies, implement changes, and accept or reject based on statistical comparison.

Instead of building a custom agent from scratch, you give Claude Code the right environment — MCP servers, a meta-knowledge library, SDK-aware plugins, performance profiles — and let it figure out what works. The innovation is the environment, not a new agent architecture.

The Approach

Most agent optimization tools build custom AI systems. AutoPerf takes a different path: curate the right _environment_ for Claude Code and let it operate autonomously.

Each iteration agent runs in an isolated git worktree with:

Context7 — live documentation for Vercel AI SDK, Google ADK, LangChain, and Claude Agent SDK. Current docs, not stale training data.
Nia — semantic search and deep research across inde