code-simplifier
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill does not contain any executable scripts, remote downloads, or hardcoded credentials. It functions primarily as an instructional guide for interacting with a local subagent utility.
- [INDIRECT_PROMPT_INJECTION]: The skill instructions include a data ingestion surface where external content, such as task context and issue/PR numbers, is passed to a subagent.
- Ingestion points: Data is passed to the
codex-subagenttool via a stdin heredoc as shown in theSKILL.mdrun example. - Boundary markers: The provided command example uses a quoted heredoc (
<<'EOF') which is a security best practice that prevents the local shell from interpolating the content, providing a safe boundary for the agent's input. - Capability inventory: No high-privilege capabilities, such as network access or direct file-system modification, are defined or requested within the skill's instructions.
- Sanitization: No explicit sanitization or filtering of the ingested external content is defined in the instruction file.
Audit Metadata