Approval gates for AI agents: how Sahkarmi decides what a coworker may do

Описание и применение

Sahkarmi реализует модель разрешений, где каждое действие агента классифицируется в 16 категорий, из которых шесть требуют ручного одобрения. Автоматические и однократные одобрения записываются с истечением срока, а вход в браузер осуществляется вручную, чтобы защитить учётные данные.

Задача

AI agents can perform irreversible actions (payments, deletions, form submissions) that, if hallucinated, can cause real harm; a system is needed to ensure human approval before such actions.

Результат

Sahkarmi’s design reduces approval fatigue through sensible defaults, limits risky actions to human approval, and protects credentials via a browser take‑over flow, thereby preventing unintended or malicious agent actions.

Рабочий процесс

• 1. Agent proposes a tool call.
• 2. Permission policy checks the call’s category.
• 3. If category is auto‑allow, action proceeds.
• 4. If category is auto‑deny, action is blocked.
• 5. If category is ask, a pending‑approval token is generated and an approval request is sent to the user (desktop or phone).
• 6. User reviews the request, optionally sees a diff of the effect, and approves or rejects.
• 7. Approval is recorded; delegated approvals expire after 30 days, single‑use approvals after 7 days.
• 8. Approved actions are executed; denied actions are logged and not performed.

Доказательства

The model proposes. A human approves. The decision is recorded.

Источник

Six categories are special. Payments, messages to customers or third parties, publication, shell commands, destructive file operations and official or government forms are hard‑coded to ask.

Источник

The sandboxed browser is handed to the human, who signs in with their own hands; the coworker then takes the session back. The model never sees the password or the OTP.

Источник