workflow-automation

Warn

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses 'npx claude-flow', which fetches the 'claude-flow' package from the npm registry at runtime. This package is not part of the trusted vendors list or a verified source.
  • [COMMAND_EXECUTION]: Shell commands are executed via npx to perform workflow management tasks such as creation, execution, and export.
  • [PROMPT_INJECTION]: The skill processes workflow definitions that assign tasks to various agent types (e.g., researcher, coder, tester). This is a vulnerability surface for indirect prompt injection.
  • Ingestion points: YAML workflow files and command-line arguments specifying task descriptions.
  • Boundary markers: No markers or instructions are provided to the agents to distinguish between system instructions and untrusted content within the workflow steps.
  • Capability inventory: The skill can orchestrate multiple agents and execute automated steps including code implementation and deployment flows.
  • Sanitization: There is no validation or sanitization of the 'task' strings or YAML structures before they are processed by the underlying agents.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 1, 2026, 04:33 PM