Project Optimizer

One click to find out what's missing from your agent setup — and fix it without leaving the app.

AI agents are only as good as the instructions you give them. But keeping CLAUDE.md, rules, skills, and config files complete and consistent is hard to do by hand — especially as projects grow. The project optimizer scans your project and surfaces exactly what's missing or needs attention.

It runs 35+ checks across six categories: file presence, content quality, code organization, project setup, cross-file consistency, and stack-specific recommendations. Every finding is ranked by severity, explained in plain language, and paired with a one-click fix. You can also export the full report as a prompt for Claude to act on directly.

The Explore tab gives you a directory-level view of every agent instruction file across your codebase. Browse into any folder, see which directories have CLAUDE.md or rules files, and add new ones right where they're needed — so coverage grows alongside your project.

Why it matters

Health score at a glance

A weighted score across files, content, code, and setup shows your project's agent readiness. Category-level dots reveal exactly where to focus.

One-click fixes

Each gap suggests a specific action — create a missing file, add a Build & Run section, insert a block from the library. Click to fix, and the score updates instantly.

Stack-aware detection

The analyzer reads your manifest files to detect Next.js, Python, Go, Swift, and more — then surfaces recommendations specific to your stack.

Explore your codebase

The Explore tab maps every agent instruction file across your project. Browse into any directory, see what's covered, and add new files right where they're needed.

Export for agents

Copy the full optimization report as a prompt. Paste it into Claude Code and let the agent fix every gap with project-specific content — not generic placeholders.

Ready to try it?

Create a free account and download Crystl for macOS.

get started getting started guide →

Frequently asked questions

What does the optimizer check?
35+ signals across six categories: file presence (CLAUDE.md, .claudeignore, etc.), content quality (Build & Run section, architecture docs, file size limits), code organization (files over 500 or 1000 lines), project setup (skills, rules, MCP servers, hooks), cross-file consistency (stale paths, config conflicts), and stack-specific recommendations.
How does the scoring work?
Each gap subtracts points from a base score of 100, weighted by severity. Content is weighted 3x, files and code 2x, and setup 1x. The result is classified as high (75+), medium (40-74), or low (below 40).
What can I fix directly from the optimizer?
Most gaps have a one-click action: create a missing file, add a section to CLAUDE.md, or open a file in the block editor. For everything else, the export button generates a report you can hand to Claude Code.
Does it work with monorepos?
Yes. The optimizer detects subpackages and checks for CLAUDE.md files in each one. It also checks for path-scoped rules in .claude/rules/.
Does it work over SSH?
Yes — the optimizer reads remote project files through Crystl's SSH connection. Health scores, gap detection, and file creation all work on remote machines.
How often should I run it?
After any major project change — new dependencies, restructured directories, or when onboarding a new team member. The re-analyze button refreshes instantly.

More features