litestar-styleguide

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides references to several external GitHub repositories, including official Litestar reference applications (litestar-fullstack-inertia, litestar-fullstack) and community-maintained service pattern examples (litestar-sqlstack). These are established resources within the Litestar ecosystem.
  • [EXTERNAL_DOWNLOADS]: CI/CD templates in references/ci-cd.md reference well-known GitHub Actions and pre-commit hooks, such as astral-sh/setup-uv, oven-sh/setup-bun, ruff-pre-commit, and pre-commit-hooks. These are industry-standard tools for software validation.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a shared resource for other AI agent skills, defining rules for code style, testing, and CI/CD. While this structure creates a surface where provided instructions could influence downstream agent behavior, the content is restricted to standard software development guidelines and lacks exploitable capabilities like arbitrary command execution or network exfiltration.
  • [COMMAND_EXECUTION]: Documentation for the Google Developer Knowledge MCP server in references/google-developer-knowledge-mcp.md includes setup commands for CLI tools (e.g., claude mcp add, gcloud auth). These are instructional examples for user opt-in and do not represent silent or malicious execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 07:08 PM
Security Audit — agent-trust-hub — litestar-styleguide