codex-chatgpt-web-integration
Fail
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The installation process for macOS, Linux, and Windows involves downloading scripts from a remote GitHub repository (miuuyy/codex-chatgpt-web) and piping them directly into a shell interpreter (sh) or PowerShell (iex). This allows an unvetted remote script to execute arbitrary code on the host machine without user inspection.
- [COMMAND_EXECUTION]: The skill facilitates the execution of local shell commands and filesystem operations through a 'Full Harness Mode' using the Model Context Protocol (MCP). This enables an external AI model to initiate commands on the user's system.
- [EXTERNAL_DOWNLOADS]: The skill relies on binaries and scripts hosted on a repository that is not affiliated with a trusted vendor or the skill author's verified infrastructure.
- [DATA_EXFILTRATION]: The integration requires setting up an OpenAI tunnel and providing API keys, creating a bi-directional network channel. When combined with the skill's ability to read local files via MCP tools, this presents a significant risk for data exfiltration.
Recommendations
- HIGH: Downloads and executes remote code from: https://github.com/miuuyy/codex-chatgpt-web/releases/latest/download/install-launcher.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata