frontend-design-extractor

Fail

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to run shell scripts (scripts/scan_ui_sources.sh and scripts/generate_output_skeleton.sh) which are not included in the provided file set. These scripts are invoked with user-supplied arguments such as repository roots and output paths, which could be exploited if the scripts are malicious or missing.
  • [REMOTE_CODE_EXECUTION]: The shell scripts referenced (scripts/scan_ui_sources.sh and scripts/generate_output_skeleton.sh) are missing from the skill package. Directing an agent to execute external, unverifiable scripts at runtime is a high-risk pattern.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external frontend repositories which may contain malicious instructions designed to influence the agent.
  • Ingestion points: Frontend repository source code (React, Vue, Angular, etc.) accessed during the source inventory phase (SKILL.md, Workflow Step 1).
  • Boundary markers: Absent. There are no instructions for the agent to use delimiters or to ignore instructions embedded within the codebase being scanned.
  • Capability inventory: The agent is granted the ability to execute shell scripts, write to the local filesystem (ui-ux-spec/ folder), and modify existing project files during the refactor flow.
  • Sanitization: Absent. The skill does not specify any sanitization, filtering, or validation of the content read from the target repository.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Aug 19, 2026, 07:46 PM
Security Audit — agent-trust-hub — frontend-design-extractor