x9-codex-delegation

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute the codex CLI and several POSIX shell utilities (mktemp, chmod, umask, trap, git) to manage the lifecycle of delegation prompts and verify project state.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests codebase content and user instructions to pass to an external executor, creating a surface for indirect prompt injection if malicious content exists in the processed files.
  • Ingestion points: The skill reads project files and user briefs to populate a temporary transport file (DELEGATION_PROMPT_FILE) defined in references/codex-cli.md.
  • Boundary markers: Instructions recommend the use of --disable hooks and --ephemeral flags to isolate the execution context.
  • Capability inventory: The skill uses subprocess calls to codex and local shell commands; it writes logs to ~/.local/share/x9/ and temporary files to /tmp.
  • Sanitization: The skill employs strict security hardening for data transport, including umask 077 and chmod 600 for temporary files, and passes prompts via stdin to prevent shell injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 07:34 AM