Frequently Asked Questions
Find answers to common questions about Comper
How is Comper different from code search or APM platforms?
Comper creates a living, explorable map of your entire software system, not just search results or runtime metrics. It shows relationships, ownership, evolution, and risk across your codebase in one unified view, grounded in actual code.
Can AI agents use Comper?
Yes. The Comper MCP works with any MCP-compatible agent and gives it the full public API, read and write. Agents can list and filter repositories, read summaries and facts, search code across the org, check pull requests for secrets and suggested reviewers, render C4 diagrams, and act on what they find, such as creating or moving Jira issues. Write operations need a personal access token with write scope.
How do you handle source code security?
Comper runs on-premises, in your cloud, or we also have our own cloud. Your code never leaves your infrastructure. We support SSO, encrypted storage, and you can bring your own LLM for AI features. Security is built in, not bolted on.
What does setup look like?
Our deployment is kept simple. A single docker image + a database. Once your board is up and running, connect your repositories via your Git system's API token (GitHub, GitLab, BitBucket, or Azure DevOps). Comper analyzes git history and generates your system map automatically. Most teams see their map within hours, not weeks.
Can we start with one team or domain and expand?
Yes. Start with a single repository or team's codebase. As you add more repos, Comper builds a unified view. There's no need to map everything at once. Expand at your own pace.
How do diagrams stay current as code changes?
Comper continuously syncs with your repositories. The map updates automatically as code evolves, so you're always looking at the current state. Not a snapshot from last month.