Code Lexica does two things: it shows you what your team actually spends on AI coding, and then it cuts that bill without changing how anyone works.
Both halves run off one small CLI (cdlx) that sits next to the coding tools your team already uses: Claude Code, Codex, and Cursor.
Track: see the bill (free)
Track meters real usage across your team's coding tools and shows it in one dashboard: tokens, models, True Cost at API rates, and breakdowns by member, model, repository, and tool. It's free forever, with unlimited team members, and setup is about 5 minutes per developer.
- Team dashboard with usage and True Cost, updated automatically
- Weekly and monthly briefings that explain what changed and why
- A team leaderboard you can share org-wide, no accounts needed
Optimize: cut it (free 21-day trial)
Optimize runs as a lightweight sidecar next to each developer's agent, steering model, effort, and context decisions in the background. Typical token-spend reduction is up to 30% without access to your codebase, and 30–60% with your codebase indexed.
You don't have to take our word for it: a paired A/B test runs the same task with and without Code Lexica on your own repository and scores both arms on tokens, cost, duration, and pass rate.
Where to go next
Start with "Getting started with Track" to get the dashboard filling in, then "Starting your Optimize trial" when you're ready to cut the bill.