x9-codex-delegation

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides structured templates and shell recipes for the agent to invoke the codex command-line tool. It distinguishes between one-shot execution and session resumption, providing specific instructions for read-only and write-capable sandbox modes.
  • [DATA_EXPOSURE]: The skill is designed to interact with local configuration at ~/.codex/config.toml and maintain an operational log at ~/.local/share/x9/codex-delegation-log.md. It includes instructions to avoid logging sensitive data, repository names, or personal paths.
  • [SAFE]: The skill implements robust security controls for handling temporary delegation data. It uses mktemp with restricted permissions (umask 077, chmod 600) and shell trap signals to ensure that transport files containing prompt content are securely created and deleted after use.
  • [SAFE]: The instructions mandate a verification workflow for write-capable tasks, requiring the agent to inspect git diff and run local verification tests. This serves as a significant mitigation against potential indirect prompt injection attacks originating from processed project files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 07:41 AM
Security Audit — agent-trust-hub — x9-codex-delegation