Industry Newsagentsprompt injectionanthropicopen source
Hacker Installs OpenClaw Via Cline Vulnerability
7.1
A hacker exploited a prompt-injection vulnerability in the open-source AI coding agent Cline to install the OpenClaw agent on users' computers, The Verge reports. Security researcher Adnan Khan had published a proof-of-concept days earlier; the exploit abused Anthropic's Claude-based workflow and was fixed only after public disclosure. The incident underscores growing security risks from autonomous AI agents.
Key Points
- 1Exploited Cline's prompt-injection vulnerability to install OpenClaw on users' computers.
- 2Highlights that prompt injections can subvert Claude-powered workflows and enable unauthorized actions at scale.
- 3Urges developers to fix reported vulnerabilities and restrict autonomous agent capabilities or adopt lockdown measures.
Scoring Rationale
Timely, concrete exploit reporting drives score; limited technical detail and single-source reporting reduce overall certainty.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
