237K+ AI events recorded on our own fleet · last 30 days AI events on our own fleet · 30d
All articlesCompany

Introducing Tracelet: visibility and control for AI coding tools

AI coding tools are already on your developers' machines. Tracelet shows what they do, applies the rules you set, and gives security, engineering and leadership one shared view.

AI coding tools have moved faster than the policies meant to govern them. Assistants and agents now read repositories, run commands and call external services on developer machines every day. For most organisations, the honest answer to “what are our AI tools doing?” is still “we’re not sure.”

We built Tracelet to change that answer, without slowing developers down.

What Tracelet does

Tracelet is a control plane for AI coding tools. It runs on enrolled machines, alongside the tools your teams already use, and does three things.

  • Shows what is happening. Which AI tools are in use, by whom, and which actions they take on real machines and repositories.
  • Applies the rules you set. Policies cover supported actions, like reading private keys or sending data to an unapproved service. Each decision is explained to the developer, with the approved path when there is one.
  • Keeps a decision record. Every allow, guide or block is recorded with its context, so security, engineering and leadership can review the same facts.

Observe first, then set rules

We think governance should start with evidence, not guesswork. Tracelet begins by observing, so you can see the real landscape before you restrict anything. Most AI usage turns out to be ordinary and productive. Rules can then focus on the few actions that carry real risk.

Guide in flow, block when it counts.

Built to stay in your environment

Decision records are sensitive, so Tracelet is built to keep them close. The control plane runs single tenant, in your cloud or data centre. Decision context is recorded by default; content capture is opt-in and encrypted, with retention and access that you set. Read more on our security and privacy page.

One view for every team

  • Security gets a clear view of AI activity, the policies that apply and the decisions that need review. See security and governance.
  • Engineering gets AI tools with clear rules, and a fleet view of which machines are covered. See engineering.
  • Leadership gets adoption, value and risk in one place. See leadership.

Work with us

Tracelet is built and operated by Insighture. We are working closely with a small group of design partners to shape what comes next. If you want to adopt AI tools quickly while keeping your business in control, become a design partner.

See what your AI tools are doing. Keep your business in control.

Let’s talk