create-rule
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The workflow in SKILL.md generates shell commands using placeholders like
<test-compiled>and<name>(e.g.,opentaint test rule run <test-compiled>/sinks). These inputs are not explicitly sanitized, creating a risk of command injection if the strings contain shell metacharacters.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by using external inputs to author detection rules.\n - Ingestion points: Data from
<requirements>and<test-project>files is used to derive rule patterns.\n - Boundary markers: The instructions do not specify any delimiters or safety warnings to handle malicious content within requirements or test projects.\n
- Capability inventory: The skill is capable of file system writes to rules directories and executing the
opentaintCLI for testing and scanning.\n - Sanitization: No input validation or shell-escaping logic is described for the variables used in commands or for the content derived from external requirements.
Audit Metadata