Skills
skills/petekp/claude-code-setup/codex/Gen Agent Trust Hub

codex

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection by forwarding untrusted data to an autonomous executor.
  • Ingestion points: User input following the /codex command in SKILL.md.
  • Boundary markers: Absent; instructions specify that the prompt is preserved verbatim.
  • Capability inventory: The tool runs with --full-auto (enabling file system modifications) and supports --search for web access within scripts/run-codex.sh.
  • Sanitization: Absent; the user input is piped directly to the binary's stdin without filtering or escaping.
  • [COMMAND_EXECUTION]: The script scripts/run-codex.sh executes the codex binary from the system PATH to perform requested tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 01:27 PM