{"data":{"id":"67ee65d3-d868-4642-aaef-ff6022f1a933","title":"GenAI and Agentic AI Exploit Roundup Q2 2026","summary":"A quarterly roundup of six GenAI and agentic AI cases, covering April 1 through June 30 2026, shows how untrusted content and delegated access create security exposure. The selection is not exhaustive. Two cases involve observed malicious activity and four involve research demonstrations or controlled validation. The Comment and Control entry, disclosed April 15 2026, describes attacker-controlled GitHub text redirecting Anthropic Claude Code Security Review, Google Gemini CLI Action and GitHub Copilot Agent deployments into exposing runtime secrets. Exposure depended on workflow permissions and available secrets.","solution":"Defenders should constrain tool execution and credentials, preserve input provenance, and monitor the actions agents take. Controls should validate each action outside the model and record the identity, tool, input source, and destination.","labels":["security","research"],"sourceUrl":"https://genai.owasp.org/2026/07/08/genai-and-agentic-ai-exploit-roundup-q2-2026/?utm_source=rss&utm_medium=rss&utm_campaign=genai-and-agentic-ai-exploit-roundup-q2-2026","publishedAt":"2026-07-09T04:38:14.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"medium","attackType":["prompt_injection","data_extraction"],"issueType":"research","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":["Google","Microsoft","Anthropic"],"affectedVendorsRaw":["Claude Code Security Review","Gemini CLI Action","GitHub Copilot Agent","Anthropic Claude Code","Google Gemini CLI","GitHub Copilot"],"classifierModel":"claude-haiku-5-5","classifierPromptVersion":"v4","summaryPromptVersion":"v2","headline":null,"headlinePromptVersion":null,"cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":null,"epssScore":null,"epssCheckedAt":null,"kevDateAdded":null,"advisoryAliases":null,"affectedPackagesSource":null,"affectedPackagesCheckedAt":null,"patchAvailable":null,"disclosureDate":"2026-07-09T04:38:14.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"moderate","impactType":["confidentiality","integrity"],"aiComponentTargeted":"agent","llmSpecific":true,"classifierConfidence":0.8,"researchCategory":"industry","atlasIds":["AML.T0051"]}}