cubrid-ci-trigger
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill interpolates user-provided PR identifiers directly into shell commands, such as
gh pr view <pr-url-or-number-if-given>. This pattern relies on the execution environment's sanitization of the$ARGUMENTSvariable to prevent command injection. - [PROMPT_INJECTION]: The skill processes untrusted external data by reading PR comments to verify if a CI run is already pending. This creates an indirect prompt injection surface. Ingestion points: External GitHub PR comments retrieved in SKILL.md (Step 3). Boundary markers: None. Capability inventory: PR commenting via
gh, local repository status viagit. Sanitization: The skill uses basic regex filtering on comment bodies but does not escape the retrieved content before it enters the agent's context.
Audit Metadata