{"data":{"id":"04973db5-6b4e-46d2-b04d-73c08326e26b","title":"Agent Commander: Promptware-Powered Command and Control","summary":"Promptware-powered command and control (C2, a system attackers use to remotely control compromised devices) refers to using prompt injection (tricking an AI by hiding instructions in its input) attacks against AI tools like ChatGPT to create a malicious control channel. Researchers have demonstrated that by combining features like browsing and memory capabilities in AI systems, attackers can build complex, malware-like prompt injection payloads that function similarly to traditional malware for remote control purposes.","solution":"N/A -- no mitigation discussed in source.","labels":["security","research"],"sourceUrl":"https://embracethered.com/blog/posts/2026/agent-commander-your-agent-works-for-me-now/","publishedAt":"2026-03-17T03:20:58.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":["prompt_injection"],"issueType":"news","affectedPackages":null,"affectedVendors":["OpenAI"],"affectedVendorsRaw":["ChatGPT"],"classifierModel":"claude-haiku-4-5-20251001","classifierPromptVersion":"v3","cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":null,"epssScore":null,"patchAvailable":null,"disclosureDate":"2026-03-17T03:20:58.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"moderate","impactType":["integrity","availability"],"aiComponentTargeted":"api","llmSpecific":true,"classifierConfidence":0.75,"researchCategory":null,"atlasIds":null}}