php
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted user code and project files while possessing the capability to execute local binaries. Malicious instructions within the code being analyzed could influence the agent's behavior during the review or testing phases.
- Ingestion points: Processes user-provided PHP source files, test files, and project configuration files (e.g.,
phpunit.xml,phpstan.neon). - Boundary markers: The instructions lack specific delimiters or instructions to treat ingested code as untrusted data.
- Capability inventory: The agent is directed to execute shell commands for testing (
phpunit) and analysis (phpstan,psalm,rector,php-cs-fixer) using binaries located in the project's./vendor/bin/directory. - Sanitization: No sanitization or validation of the input code is defined before processing or command execution.
- [COMMAND_EXECUTION]: The skill explicitly directs the agent to execute command-line tools for development tasks, such as running the PHPUnit test suite and various static analysis tools. While these are standard development workflows, the execution of these commands on potentially untrusted code enables the identified indirect prompt injection surface.
Audit Metadata