Getting Started

IntroductionLocal ProxySDK Quick Start

AI ROI

Analytics & ReportsTeam ManagementGitHub Integration

Cost Management

Smart RoutingBudget & AlertsPrompt ClassifierCaching & Performance

Integrations

OpenAIAnthropicMCP ServerSlack

Reference

SDKREST APIErrors
Docs/Getting Started/Introduction

Documentation

Measure AI ROI for your engineering team. Track how AI tools impact cycle time, throughput, and code quality.

Get Started — 5 minutes, no code changes

Recommended

Install the local proxy to automatically track AI usage from Claude Code, Cursor, and Kiro.

1. Install

brew tap costlens-dev/tap
brew install costlens

2. Setup (auto-configures your shell)

costlens setup

3. Start

costlens start
Full proxy guide →

Install GitHub App

Install the CostLens GitHub App on your org to track commits, PRs, and cycle time. We never modify your code.

Install GitHub App →

AI ROI Measurement

Local Proxy

Track AI usage from any coding tool — no code changes

AI Impact Reports

Weekly reports comparing AI-assisted vs manual work

Confidence Scoring

How we detect AI-assisted commits (methodology)

GitHub Integration

Automatic PR lifecycle tracking and baseline comparison

Cost Management

Smart Routing

Route simple prompts to cheaper models, save 20-40%

Budget Alerts

Set limits, get notified via Slack before overspending

Cost Analytics

Per-model, per-developer, per-feature cost breakdown

Advanced

SDK Integration

Per-prompt attribution for apps that call LLM APIs directly

MCP Server

Query cost data from inside Claude Code, Kiro, or Cursor

REST API

Direct API integration for any language

Slack Integration

Budget alerts and weekly digests in Slack

Reference

OpenAI

GPT-4o, GPT-4.1, GPT-4.1-nano

Anthropic

Claude Sonnet 4, Opus 4, Haiku

Error Handling

Error codes and troubleshooting

Next

Local Proxy →