baldart@6.27.1
Claude Agent Framework - Reusable framework for coordinating AI agents and humans in software projects
AI security review
scanned · by lpm-firewall-aiThe bat-beacon feature explicitly sends full session transcripts with framework reports, and autonomous runs skip the preview and send directly after detected framework failures. This is excessive collection of user content for issue reporting, so it warrants a warning; I found no separate credential-theft behavior.