skills/johannhipp/skills/coss-svelte/Gen Agent Trust Hub

coss-svelte

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured guidance and reference materials for the coss-svelte UI component library, mapping COSS UI patterns onto Svelte 5 primitives.
  • [EXTERNAL_DOWNLOADS]: Documents the installation of the coss-svelte (vendor resource) and bits-ui (well-known behavioral primitive library) packages using standard package managers.
  • [COMMAND_EXECUTION]: Includes documentation for standard project maintenance commands, such as linting with Biome, type-checking, and running tests within the development workflow.
  • [PROMPT_INJECTION]: Evaluated the surface for indirect prompt injection from processed data.
  • Ingestion points: The skill processes user-provided Svelte and SvelteKit source files to provide implementation guidance and component migration advice.
  • Boundary markers: No explicit delimiters or instructions for separating user code from agent instructions are defined in the skill files.
  • Capability inventory: The skill defines component usage patterns but does not request or define autonomous network access, sensitive file system operations, or dangerous system calls.
  • Sanitization: The skill relies on the agent's default processing and does not specify custom sanitization or validation for processed code snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 11:39 PM
Security Audit — agent-trust-hub — coss-svelte