frontend-api

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • PROMPT_INJECTION (HIGH): The skill is susceptible to Indirect Prompt Injection (Category 8). It is designed to ingest data from external API endpoints (SKILL.md) to implement fetching logic. There are no boundary markers or instructions to sanitize or ignore embedded instructions within the API responses. Because the skill has Bash and Write capabilities, a malicious API response could manipulate the agent into executing system commands or injecting backdoors into the src/api/ or src/services/ directories.
  • COMMAND_EXECUTION (MEDIUM): The skill is explicitly granted Bash tool permissions in its metadata. While this is likely intended for development tasks, the presence of shell access in a skill designed to interact with external web content significantly elevates the impact of any potential injection attack.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 01:13 PM