Works inside Claude Code
No browser tab, no upload flow. Arclat runs as an MCP server inside the editor you already live in.
MCP · Claude Code
Your AI agent is slow.
Arclat tells you exactly why, and fixes it.
Arclat audits your agent codebase for latency bottlenecks, ranks them by the time they cost, and applies fixes, all from a single Claude Code conversation. No log files, no context switching.
1 line
Add to Claude Code config and you’re running.
Ranked
Bottlenecks sorted by time lost, not guesswork.
Auto-fix
Apply patches directly from Claude Code.
One-line install
Add to Claude Code config
{
"mcpServers": {
"arclat": {
"command": "uvx",
"args": ["arclat-mcp"],
"env": {
"ARCLAT_API_KEY": "your_api_key_here"
}
}
}
}Then say in Claude Code:
“Audit this codebase for latency issues”
Latency profiling that lives in your editor
Everything Arclat does happens where you already work. Point it at a codebase and it does the rest.
Reads your actual code
AST-based analysis on your real .py source, not logs, not traces, not guesses. The true source of bottlenecks.
Ranked by impact
Findings sorted by the time they cost, so you fix what actually matters instead of chasing micro-optimizations.
One-command fixes
Ask Claude Code to apply a fix and Arclat patches the file in place, with a diff you approve first.
CI-ready
Wire the same engine into a GitHub Action and fail the build when a change regresses agent latency.
Your code stays yours
Analysis runs locally through the MCP server. Only usage metadata syncs, never your source.
From install to fix in under a minute
01
Add one config block
Drop the Arclat MCP entry into your Claude Code config and paste your API key.
02
Ask for an audit
“Audit this codebase for latency issues.” Arclat parses your code and ranks every bottleneck.
03
Apply the fixes
Review the ranked findings, then let Arclat patch the ones you approve, right in your editor.
Simple pricing
Start free. Upgrade when you want unlimited audits and in-editor fixes.
Stop guessing where the latency is.
Install Arclat in Claude Code and get a ranked latency audit of your agent codebase in under a minute.