crewai-tools’ MCPServerAdapter, which spawns the server over stdio (or connects over SSE/HTTP) and turns its tools into CrewAI Tool objects.
1
Install
npx -y @xentfi/mcp-server as the command instead — slightly slower to start, no PATH setup required.2
Wire it into a Crew
Letting an agent actually pay
Give a CrewAI agent thexentfi_create_payment tool only if you intend for it to move funds. Because that tool requires confirm: true, a good pattern is a two-agent crew:
- Planner agent — decides what payment should happen (recipient, asset, amount) and writes it out, without payment tools.
- Executor agent — has only
xentfi_create_payment(filter the tool list fromxentfi_toolsto just this one) and a task instructing it to call it withconfirm: trueonly if the planner’s output passed your own validation step.
MCPServerAdapter in newer crewai-tools versions also supports connecting to a remote server over SSE/Streamable HTTP if you’d rather run xentfi-mcp-http centrally and point every Crew at it — see Remote HTTP Deployment and pass the Authorization: Bearer <agent-api-key> header via that adapter’s HTTP params.xentfi_get_agent_info, etc.) — reference them by name if you filter/allowlist tools per agent.
Next steps
Tools Reference
See every available tool.
Remote HTTP Deployment
Run one shared XentFi MCP endpoint for every Crew.

