MCP server
MCP (Model Context Protocol) servers expose tools to agents over a standard protocol. The Fuze MCP host (@fuze-ai/agent-mcp) intercepts the transport so every MCP tool call flows through the same evidence pipeline as a native tool.
What you'll build: a filesystem MCP server attached to your agent with fingerprint pinning, tool allowlist, and audit spans. Prerequisites: Cerbos policies, MCP admission re-uses the same engine. Next: verify the audit chain end to end.
Install
Wire the filesystem MCP server
Admission policy
The MCP host enforces an admission policy via Cerbos (mcp.admission.yaml). Servers not on the fingerprint allowlist are rejected at connect time. Tools not on the per-server allowlist are rejected at dispatch.
A non-isolated sandbox tier on an MCP server is refused.
Use MCP tools in your agent
The host exposes MCP tools as FuzeTool instances; pass them directly to defineAgent:
Each MCP-backed tool's threat boundary is derived from the server's declared capabilities and the host's policy.
Cap the tool-list token budget
Cap the size of tools/list results to keep prompt tokens predictable:
When the budget is exceeded, the host emits a fuze.mcp.budget warning span and returns a truncated tool list. A hard cap fails closed.
Evidence
Every MCP dispatch emits both fuze.tool (as for native tools) and fuze.mcp spans: