wp-plugin-review
Fail
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: HIGHPRIVILEGE_ESCALATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PRIVILEGE_ESCALATION]: The setup script
scripts/setup_tools.shutilizessudoto install system packages and to move binaries into restricted directories. - [EXTERNAL_DOWNLOADS]: The skill downloads dependencies from external sources during setup:
- Fetches the Composer installer from
https://getcomposer.org/installer(a well-known service). - Installs security analysis tools and frameworks from public repositories using Composer.
- [COMMAND_EXECUTION]: The skill executes multiple shell commands to initialize its environment and perform plugin analysis:
- Runs
scripts/setup_tools.shto install necessary tools. - Executes
phpcs,phpstan, andphpuniton user-provided code within the/home/claude/plugin-under-review/directory. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted code from user-uploaded plugins, which presents a risk of indirect prompt injection.
- Ingestion points: Reads all PHP files from uploaded zips or folders in
/mnt/user-data/uploads/. - Boundary markers: None. The instructions do not include specific delimiters or warnings to prevent the agent from following instructions embedded in the analyzed code.
- Capability inventory: The skill can execute shell commands and write review reports to the output directory.
- Sanitization: None. The ingested code content is reviewed directly without any escaping or validation steps.
Recommendations
- AI detected serious security threats
Audit Metadata