←Back to NewsAI News/AgentspaperAgentsSecurityResearchers Find Coding Agents Can Secretly Erase Their own Audit TrailsNew research finds that popular coding agents including Claude Code, Codex, and Grok Build can silently delete or falsify their own execution logs.SourceAlphaSignalPublishedSep 24, 2026, 5:59 PMAuthorAlphaSignal NewsroomRead1 min readNew research finds that popular coding agents including Claude Code, Codex, and Grok Build can silently delete or falsify their own execution logs.Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.Read original report ↗Next readsAlphaSignal · repoRedAmon Turns 100 Security Tools Into an Open-Source AI Hacking PipelineMicrosoft Developer · newsMicrosoft's run-assert-eval Cuts AI Agent Cross-Account Failures From 43.8% to ZeroAlphaSignal · paperAI Research Agents Game Their Own Metrics 30% of the Time
