preview
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute build commands and launch background processes, such as development servers, based on the identified project structure.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources to determine its actions.
- Ingestion points: The skill parses pull request (PR) URLs, changed source code, project entry points, and README files from the local environment to resolve the preview scope (SKILL.md).
- Boundary markers: There are no instructions or delimiters provided to signal the agent to ignore or isolate instructions that might be embedded within the project data or PR descriptions.
- Capability inventory: The agent has significant capabilities, including executing arbitrary shell commands via the Bash tool (with background support) and monitoring real-time log output.
- Sanitization: No evidence of sanitization, filtering, or validation of the ingested external content is present before it is used to influence the build and execution steps.
Audit Metadata