Context Data

Context Data makes InterLock, a proxy that sits between AI agents and the databases, files and APIs they use in production. The idea is that agent platforms control which tools an agent may call, while nothing controls which rows those tools may read or change. InterLock works at that data layer, on protocols the data already speaks: MCP, the Postgres wire protocol and HTTP. Each agent receives its own identity and key. Roles allow or deny actions per source and, for SQL sources, per table and column, with deny as the default. SQL is parsed before it runs and anything a role does not permit is refused. Personal data in results, such as national ID numbers, is detected and masked before the agent sees it, and policies can also redact named columns, rate limit an agent or cap the risk of its writes. Writes are classified by risk: routine inserts pass, while updates, unbounded deletes and schema changes wait in a queue for a person to approve or reject in a console, with an optional Slack notification. Every request is written to an audit log. It is built for engineering and security teams putting agents in front of sensitive production data. Agents and data stores need no changes. The site links a quick start and a GitHub repository. The vendor does not publish prices. Enquiries go through a contact form on the site.