solid-impl-solidstart

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No attempts to override agent behavior or bypass safety filters were detected in any of the files, including SKILL.md and its references. The instructions maintain a strictly technical and instructional tone.
  • [DATA_EXFILTRATION]: No sensitive file paths, credential harvesting patterns, or unauthorized network operations were identified. Code examples consistently use generic placeholders like 'db.getUser' and focus on standard application logic.
  • [REMOTE_CODE_EXECUTION]: No remote script execution patterns (e.g., curl piped to shell) or instructions to install untrusted external dependencies were found. All external links point to the official solidjs.com documentation domain.
  • [COMMAND_EXECUTION]: The skill does not instruct the agent to execute dangerous shell commands or acquire elevated privileges (e.g., sudo, chmod).
  • [SAFE]: All analyzed components follow established web development best practices for the SolidStart framework and do not exhibit any malicious intent or obfuscation techniques.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 09:18 AM
Security Audit — agent-trust-hub — solid-impl-solidstart