agent-browser

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's examples and workflow use direct CLI fill commands with plaintext values (e.g., agent-browser fill @e2 "password123"), which requires the agent to insert secret/password strings verbatim into its generated commands and outputs.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to navigate and snapshot arbitrary public webpages (e.g., "agent-browser open " and the example "agent-browser open https://news.ycombinator.com" with "snapshot -i --json"), so the agent ingests untrusted, user-generated web content and uses that parsed content to decide clicks and other actions.