FAQ

Hakama adds a governance control layer to the AI agents workflow. Hakama govenrns the scoped work, checks agent output and write attempts, and preserves telemtry and receipts for developer and team review.

Hakama is designed for engineering teams that want to govern AI coding agents. Solo developers can also use it to keep agent work scoped and reviewable.

A pilot gives your team hands-on access to the Hakama ecosystem: governed agent launches, scoped work, receipts, evidence trails, and review surfaces for developers and non-developers.

No. Hakama makes code review easier by recording intent, constraints, validation results, agent activity, and supporting evidence before reviewers make decisions.

Hakama checks scoped contracts, write attempts, agent response evidence, plugin outputs, validation gates, and delivery or telemetry requirements depending on the run.

Hakama records repo-local receipts and artifacts for the request/intake, scoped and execution contracts, validation results, and supporting evidence hashes tied back to the run or change.

With our Hakama system services connected, teams can publish and review governance feeds, raw receipts, plugin evidence, QA signals, approval decisions, and operator audit trails.

Hakama currently wraps Claude, Codex, and Gemini for governed launches. Additional agent support should be described as roadmap unless it has shipped.

Hakama runs as a local governance layer. In normal use it doesn't need your source code, prompts, AI outputs, git history, branch names, commit messages, repository contents, or working-tree files.

If you choose to send us a diagnostic bundle, support file, or pilot report, that's an explicit step you control and scope. We don't do this in the background.

Hakama does not control what your AI agent does with your code, that is outside of our scope of control.

Not today, and we won't claim otherwise. If your team needs formal compliance evidence, ask us for the current security packet, data-flow description, processor list, and access-control model.

If you work in a regulated or security-sensitive environment, the best place to start is a scoped pilot. We can work with you inspect the real evidence model before any broader rollout.

Hakama is early, so we don't publish logos, testimonials, or case studies yet. We won't invent them. During your evaluation the proof comes from your own pilot: a brownfield repo and workflow you choose, a controlled AI-assisted change, a visible scope and evidence trail, and a decision record showing whether the work was accepted, rejected, or promoted.

As customers approve them, we'll publish real case studies and references.