A researcher at Forever Security says a flaw dubbed ThisBash in Claude Dispatch lets a malicious browser extension send commands to Claude Desktop as if they came from the legitimate user, according to a summary of a November 5 report in The Register.
Gal Weizman found that Dispatch’s interface and API are reachable from an ordinary logged-in claude.ai session. An extension with claude.ai access could therefore run code on the user’s machine, or on another machine signed into the same Claude account, with minimal user interaction and no browser exploit.
Anthropic reportedly acknowledged the report as valid in May but closed it as a duplicate of a flaw it was already tracking, issuing no bounty or CVE. It did not answer The Register’s questions, and no fix has been reported. Users can reduce exposure by removing unneeded browser extensions.