claude-wireframe-skill
Warn
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the agent to download and install content from an unverified GitHub repository (https://github.com/Magdoub/claude-wireframe-skill) using the
npx skills addcommand. - [COMMAND_EXECUTION]: The skill uses shell commands for environment detection and installation, specifically
[ -e ~/.agents/skills/claude-wireframe-skill ]andnpx skills add. - [INDIRECT_PROMPT_INJECTION]: The skill performs a comprehensive scan of the local codebase to extract design context, which creates an attack surface for indirect prompt injection from untrusted local files.
- Ingestion points: Full codebase scan for design patterns, colors, and components on the first run as described in SKILL.md.
- Boundary markers: No delimiters or safety instructions are provided to the agent to ignore potentially malicious content within the scanned codebase.
- Capability inventory: The skill possesses file writing capabilities (creating the
wireframe/directory), execution of parallel agents, and shell command execution. - Sanitization: There is no evidence of sanitization, filtering, or validation of the content extracted from the codebase before it is processed by the agent.
Audit Metadata