Root cause

Tool Misuse

19 documented incidents where tool misuse was the underlying cause.

What this cause produces

Destructive Action9
Security Vulnerability8
Infinite Loop1
Logic Error1

Gemini CLI silently executed arbitrary code from an untrusted repo (CVE-2026-12537, CVSS 10.0)Gemini Cli

Malicious cloned repository triggered code execution in Cursor on WindowsCursor

GPT-5.6-Sol 'accidentally deleted almost ALL' of a tester's Mac files during OpenAI's Ultra mode trialGpt 5.6 Sol

Clinejection: an AI issue-triage workflow enabled arbitrary code execution on the CI runnerCline

Claude Code ran rm -rf from the filesystem root, destroying a developer's home directory (GitHub #10077)Claude Code

CamoLeak: hidden prompt injection turned GitHub Copilot Chat into a silent code/secret exfiltration channel (CVSS 9.6)Github Copilot

Gemini CLI destroyed a user's project files after a failed mkdir, then confessed 'gross incompetence'Gemini Cli

Hacker slipped a data-wiping prompt into Amazon Q's VS Code extension, shipped to ~1M installsAmazon Q

EchoLeak: a zero-click email silently exfiltrated data from Microsoft 365 Copilot (CVE-2025-32711, CVSS 9.3)Microsoft Copilot

A hidden comment made GitLab Duo leak private source code and inject rogue HTMLGitlab Duo

Slack AI could be tricked into leaking private-channel data via indirect prompt injectionSlack Ai

Claude Code ran drizzle-kit push --force against production, wiping 60+ tables of trading data — the second such wipe in 11 daysClaude Code

Claude Code ran prisma db push --force-reset on production, dropping all 87 tables instead of a safe schema changeClaude Code

Claude Code ran git filter-repo and force-pushed mid-outage, deleting production files without the approval CLAUDE.md required (GitHub #45893)Claude Code

Gemini CLI ran git commit --no-verify against explicit instructions, then git reset --hard wiped a sprint's worth of unstaged workGemini Cli

Claude (via OpenCode) followed an error message's suggested escalation straight to `bd init --force`, wiping a Dolt-backed issue tracker's entire historyClaude

Devin CI workflow caused 836-comment spam storm on single PRDevin

Cline's execute_command reported a failing Ruff lint check as passing over Remote-SSHCline

Copilot CLI destroyed its own 233MB session log trying to "back it up" with a hardlink instead of a copyGithub Copilot