Vinay Thakker outlines AI coding agent identity and command policies

AI agents should run under their own identity, with human identity attached for tasks for one person, using allowlist command policies and credential proxies to prevent data leaks.

In an interview, Vinay Thakker, CTO of KloudStax, explains how to set identity, credentials, and network limits for AI agents. Agents should run under their own identity, with the person’s identity attached when acting on someone’s behalf. The deciding question is whether the agent is doing a job for the business or a task for one person. If it’s doing a job for the business, it gets its own identity, with permissions scoped to that job. If it’s doing a task for one person, it still gets its own identity, but each request also carries the identity of the person who asked. Borrowing the person’s login outright is ruled out because the agent’s actions become indistinguishable from the person’s. For credential proxies, the proxy verifies the agent’s identity, obtains a short-lived credential scoped to that call, and attaches it on the way out. For command policy for a coding agent, the first ten lines should be an allowlist of approved commands and restrictions including working inside one checked-out repository, running only approved commands, no credentials in environment, no opening files with secrets, network access off except for specific hosts, packages from internal repo only,…