Cloudflare gives personal agents their own traffic category, letting sites allow, limit or block each supported agent
Synopsis
Cloudflare released traffic analytics and finer-grained access policies for personal agents. Previously, sites saw these requests made on a person's behalf only as automated traffic, with little way to tell which agents were visiting or which pages they reached. Security Analytics now lists Personal Assistants as its own category and can filter by a single agent's paths, status codes and source networks; sites can also set Allow, Block on pages with ads, or Block for each supported personal agent.
Interpretation
Site owners can now identify personal agents as a separate traffic category and see how many requests each agent sends. Previously, requests to compare products, find support articles or complete a purchase on a user's behalf appeared only as automated traffic, leaving sites unable to tell whether it helped the business or created risk. The Bot Analysis chart in Security Analytics adds Personal Assistants alongside Verified Bot, Likely Human, Likely Automated and Automated, and a new Personal Assistants widget in Top statistics lists visiting agents and their request counts.
Once an agent is identified, a site can drill into that single agent's requested paths, received status codes and source networks. Without this per-agent breakdown, sites could not tell whether an agent was logging in, reading product pages, reaching checkout, or hitting errors on pages it cannot use. Filtering on one agent updates every other widget to show only that agent's traffic details.
Access control expands from a single Agents policy to separate policies for each supported personal agent. The earlier AI bot policies separated Search, Training and Agent behaviors, but Agent behavior itself had no finer options. In Security Settings, an overall Agents policy can be set, then each supported personal agent can be set to Allow (do not block), Block on pages with ads, or Block, while existing Search and Training policies stay as they are.
Perspective
The analytics and policies target site owners, security teams and fraud teams using Cloudflare, and per-agent policies cover only supported personal agents; account-level agent attribution relies on Account Abuse Protection, which is in early access.
Cloudflare says its next goal is carrying user context through an agent, helping sites assess reputation, and establishing accountability when something goes wrong; those capabilities are not yet available.