codex-plugin-cc
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing the
@openai/codexCLI tool via npm. This package is provided by a well-known organization and is hosted on a standard registry. - [COMMAND_EXECUTION]: The documentation describes using shell commands for authentication and setup, such as
!codex login. This is used to interface with the local Codex binary for user authentication. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests and processes external code content that may be untrusted.
- Ingestion points: Code diffs, uncommitted changes, and branch comparisons are read from the local repository during review or investigation tasks (found in
SKILL.md). - Boundary markers: No specific boundary markers or instructions to ignore embedded commands are mentioned in the skill description.
- Capability inventory: The skill can investigate bug reports, fix tests, and redesign components using the
/codex:rescueand/codex:reviewcommands. - Sanitization: There is no mention of sanitization or filtering applied to the code content before it is processed by the model.
Audit Metadata