field-agent-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@field-agent-mcp Do you accept HSA cards?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Forward-Deployed Agent Lab
A production-shaped, synthetic service-business agent that demonstrates the work between a model demo and a usable customer system: workflow discovery, grounded answers, tool boundaries, human escalation, evaluation, voice integration, and Kubernetes deployment.
This repository contains no client data or client code. The fictional Northstar Clinic dataset was written for reproducible testing. The operating pattern is informed by hands-on work helping small service businesses adopt AI workflows; see the anonymized case study.
System at a glance
Web / Voice / MCP client
|
v
FastAPI + MCP adapters
|
v
deterministic route + safety boundary
/ | \
knowledge intake tool human handoff
citations pending only normal / urgent
|
v
tests + synthetic evaluation + Kubernetes runtime controlsRelated MCP server: ToolBridge
What is implemented
FastAPI endpoints for agent responses, voice-ready SSML/audio, health, readiness, and evaluation
Deterministic retrieval over an approved bilingual knowledge base with source citations
Explicit escalation for emergencies, professional judgment, and ungrounded questions
MCP tools for knowledge search, intake collection, and human escalation
Local macOS text-to-speech adapter with no external data transfer
Reproducible synthetic evaluation for route, handoff, tool-selection, and grounding behavior
Compact BERT intent-router fine-tuning experiment, kept outside the deterministic safety boundary
Docker image and Kubernetes Deployment, Service, HPA, probes, resource limits, and NetworkPolicy
Automated API, core-engine, MCP-tool, evaluation, and deployment-manifest tests
Run locally
Python 3.11–3.13 is supported.
python3 -m venv .venv
source .venv/bin/activate
python -m pip install -e ".[dev]"
pytest
python scripts/run_eval.py
uvicorn field_agent_lab.api:app --reloadOpen http://127.0.0.1:8000/docs for the interactive API.
Grounded knowledge request
curl -s http://127.0.0.1:8000/v1/agent/respond \
-H 'content-type: application/json' \
-d '{"message":"Do you accept HSA cards?","locale":"en-US"}'The response includes a knowledge route and citations to the approved payment document.
Scheduling request
curl -s http://127.0.0.1:8000/v1/agent/respond \
-H 'content-type: application/json' \
-d '{"message":"I want to book an appointment","customer_id":"demo-123"}'The system returns a collect_intake tool call and explicitly says that staff must confirm availability.
Voice-ready response
curl -s http://127.0.0.1:8000/v1/voice/respond \
-H 'content-type: application/json' \
-d '{"message":"When are you open?","generate_audio":false}'Set generate_audio to true on macOS to return base64-encoded AIFF audio generated by the local say provider.
MCP server
field-agent-mcpTools:
search_service_knowledgecollect_intakeescalate_to_human
The MCP interface reuses the same core engine as the HTTP API, so safety and grounding behavior do not drift by channel.
Compact-model experiment
Install the optional training dependencies and run:
python -m pip install -e ".[train]"
python training/train_router.pyThe script fine-tunes prajjwal1/bert-tiny on four synthetic intent classes and writes transparent metrics to artifacts/router_metrics.json. The model is an experiment; deterministic rules remain the safety boundary.
Deployment
docker build -t forward-deployed-agent-lab .
kubectl apply -f deployment/k8s.yamlThe manifest intentionally includes readiness/liveness probes, non-root execution, dropped Linux capabilities, resource requests/limits, horizontal autoscaling, and a network policy.
Evidence and limitations
Generated reports are committed under artifacts/ after execution. All benchmark cases are visible in data/eval_cases.jsonl. The dataset is small and synthetic, so the reported score is a regression baseline—not a claim of clinical quality or production generalization.
Available Tools
3 toolscollect_intakeA
Collect scheduling details without claiming that an appointment is confirmed.
Inputs: name (required), service (required), preferred_time (required), callback_number (required). Returns: object
escalate_to_humanC
Queue a request for a human staff member.
Inputs: summary (required), priority. Returns: object
search_service_knowledgeB
Search approved service-business knowledge and return citations.
Inputs: query (required), locale. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: searching knowledge, collecting intake details, and escalating to a human. There is no overlap or ambiguity between them.
All tool names follow the same verb_noun pattern with lowercase and underscores: search_service_knowledge, collect_intake, escalate_to_human. This is highly consistent.
With only 3 tools, the server is well-scoped for a focused triage workflow. Each tool serves a necessary function without redundancy or bloat.
The tools cover the core lifecycle of a field agent interaction: retrieving information, gathering details, and escalating when needed. A minor gap is the lack of a confirmation or follow-up action, but this may be intentional to avoid false commitments.
Maintenance
Related MCP Connectors
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
MCP server unifying ERPs, CRMs, APIs and knowledge base for Claude, ChatGPT and Gemini.
Hosted MCP server for pre-action AI governance verdicts, model routing, and evidence verification.
41An MCP gateway for teams deploying AI tools at scale to employee-facing and headless agents.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA production-grade MCP server designed for multi-tenant, authenticated, and observable AI agent systems, enabling secure tool execution across heterogeneous data sources.68MIT
- FlicenseNot gradedqualityDmaintenanceA governed MCP server for integrating AI agents with customer data, featuring role-based access control, field redaction, and human-in-the-loop approval for secure support operations.1-
- FlicenseAqualityDmaintenanceA production-grade MCP server that provides a centralized microservice toolkit for LLM agents, enabling web search and extensible tool integration.1-
- AlicenseNot gradedqualityDmaintenanceGoverned MCP server for bank-grade agent tool access with RBAC, PII redaction, rate limiting, and audit logging.MIT