team-frontend

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its data ingestion and multi-agent coordination design. * Ingestion points: The analyst role in role-specs/analyst.md reads task descriptions, existing project files (package.json, CSS), and performs web searches. * Boundary markers: There are no explicit delimiters or 'ignore embedded instruction' warnings in the worker spawn templates defined in SKILL.md or the monitor command. * Capability inventory: The developer role can use Edit and Write tools on source code, and the coordinator can execute Bash commands. * Sanitization: No sanitization of external or intermediate data is specified before it is passed to subsequent agent phases.
  • [COMMAND_EXECUTION]: The coordinator role uses the Bash tool to perform local filesystem operations, such as creating session directory structures in roles/coordinator/role.md.
  • [EXTERNAL_DOWNLOADS]: The skill identifies a dependency on an external plugin ui-ux-pro-max-skill, providing installation commands and referencing it via local absolute search paths in specs/team-config.json.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 04:09 PM