consult-codex
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the
codexcommand-line tool usingcodex execto obtain a second opinion on code or design decisions.- [DATA_EXFILTRATION]: The skill transmits details about the project, symptoms, and hypotheses to OpenAI's Codex API for analysis. OpenAI is a well-known and trusted service provider.- [INDIRECT_PROMPT_INJECTION]: The skill ingests potentially untrusted data from the codebase or user queries and passes it to an external model. - Ingestion points: The
<question>and<long question>placeholders inSKILL.mdare intended to be populated with project background and specific technical queries. - Boundary markers: The skill does not define specific delimiters or instructions to handle malicious content embedded within the ingested text.
- Capability inventory: The skill utilizes shell command execution to interface with the
codexCLI. - Sanitization: The instructions do not specify any validation or escaping mechanisms for the data interpolated into the CLI command.
Audit Metadata