The Agent Skills Directory

[SAFE] (SAFE): Analysis of the skill instructions and provided Javascript snippets reveals no malicious intent. The skill uses standard Model Context Protocol (MCP) patterns to interface with a workflow management system.
[Indirect Prompt Injection] (LOW): The skill processes untrusted input data via the input_data parameter in workflow_execute. While this presents an ingestion point for indirect prompt injection, it is a standard functional requirement for workflow automation.
Ingestion points: input_data parameter in mcp__flow-nexus__workflow_execute (SKILL.md).
Boundary markers: Not explicitly defined in the provided snippets.
Capability inventory: workflow_create, workflow_execute (SKILL.md).
Sanitization: No explicit sanitization or validation logic is defined within the prompt instructions, relying on the underlying tool implementation.

agent-workflow