kilo-design
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a set of instructional guidelines for an AI agent to maintain frontend design consistency. It relies on internal project assets such as
DESIGN.mdandglobals.cssto define its design contract. - [COMMAND_EXECUTION]: The
AGENTS.mdfile contains maintenance instructions to execute a local script (bun run script/check-md-table-padding.ts) for validating markdown formatting. This is a standard utility for developer maintenance. - [EXTERNAL_DOWNLOADS]: The
NOTICE.mdfile contains links to the official GitHub repository for the 'Impeccable' project (github.com/pbakaus/impeccable). These URLs are provided for licensing attribution and to document the source of the adapted design references. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it is instructed to read and analyze various source code files (e.g.,
apps/web/src/components/) and then generate or modify code based on that content. - Ingestion points: Reads code from
apps/web/src/app/globals.css,apps/web/src/components/, and Storybook story files. - Boundary markers: None identified in the skill instructions to delimit untrusted code content from the agent's instructions.
- Capability inventory: The skill is capable of reading files and writing/modifying code in the repository.
- Sanitization: No explicit sanitization or validation of the ingested code content is performed before processing.
Audit Metadata