hydris-diagnostic-mcp
A security-hardened, read-only MCP server that authenticates service and user identities, enforces factory-level access grants, and audit-logs every tool call before dispatching read-only plant overview queries.
README
hydris-mcp-integration
Hydris Pulse Bridge track, Work Item 1: the Diagnostic MCP server skeleton
that will live at lite/diagnostic_mcp/ inside the Lite monorepo. It
discharges BRIDGE-2 (per-call factory authorization), BRIDGE-5
(per-call audit logging) and NFR-4 (security), and structurally
enforces BRIDGE-1 (read-only toolset).
See docs/scope.md for exactly what is and isn't built
here, and docs/decisions.md for the open-item
decisions (revocation latency, audit-sink-failure behaviour, the
non-leaking error shape, the toolset version scheme) this skeleton took a
position on, pending Nematullah's ratification.
Architecture
- Two identities, kept separate (
diagnostic_mcp/auth/): a service-principal JWT (read-only scopes, verified independently) and a per-call user-context JWT. The service token never widens access. - One pipeline for every call (
diagnostic_mcp/registry/pipeline.py): authenticate both identities, validate arguments, check the user's live factory grant (no caching), dispatch, audit-log the outcome - success, rejected, or error - every time. - Registration-time read-only gate (
diagnostic_mcp/registry/tool_registry.py): a tool withoutclassification="read_only"or without a requiredfactory_idargument is rejected when it's registered, not when it's called. - FastMCP for schema advertisement, one override for dispatch
(
diagnostic_mcp/server.py):tools/listcomes from FastMCP as usual;tools/callis intercepted once, at the low-level server, so every tool- present or future - goes through the same pipeline with no per-tool boilerplate.
- One demonstration tool,
get_plant_overview(diagnostic_mcp/tools/plant_overview.py), backed by an in-memory stub of Lite's read-service layer, seeded with the exact fixtures the source docs name (u-ops-alphagrantedfx-mbr-01;u-ops-betagrantedfx-beta-01).
Quickstart
python -m venv .venv
.venv/Scripts/pip install -e ".[dev]" # .venv/bin/pip on macOS/Linux
cp .env.example .env # then edit the two secrets
Run the server:
set -a; source .env; set +a # or export the vars another way
python -m diagnostic_mcp.server
It listens on streamable-http at http://127.0.0.1:8765/mcp by default.
Verify it end-to-end
pytest -v # includes a real streamable-HTTP client/server round trip
ruff check .
mypy diagnostic_mcp tests
python scripts/check_no_write_imports.py
To call the running server manually with the official MCP client, mint a service token and a user-context token against the same secrets the server is running with:
from diagnostic_mcp.auth.service_principal import issue_service_token
from diagnostic_mcp.auth.user_context import issue_user_context_token
service_token = issue_service_token("<LITE_MCP_SERVICE_TOKEN_SECRET>")
user_token = issue_user_context_token("<LITE_MCP_USER_CONTEXT_SECRET>", "u-ops-alpha", "org-alpha")
print(f"Authorization: Bearer {service_token}")
print(f"X-Pulse-User-Context: {user_token}")
Then call get_plant_overview with factory_id=fx-mbr-01 (granted -
succeeds) or factory_id=fx-beta-01 (not granted - denied, and both
attempts land in audit.jsonl).
Work Item 1 definition of done
| Item | Status |
|---|---|
| MCP server registered in-process, streamable-HTTP reachable | Done |
| Tool registry with read-only classification, rejecting non-conforming registration at startup | Done |
| Service principal issued with read-only scopes, scope inventory written down against the 15-tool minimum set | Done - diagnostic_mcp/auth/service_principal.py::MINIMUM_READ_SCOPES |
| Per-call user context header parsed, grants resolved, call rejected before the service token is used | Done |
| Non-leaking authorization error shape agreed and implemented | Done - see O3 in docs/decisions.md |
| Revocation latency chosen and documented | Done - see O1 in docs/decisions.md |
| Audit sink writing all six mandated fields, on success, failure and rejection | Done |
| Pulse session id propagated and recorded, for TC-BRIDGE-5.3 parity | Done - X-Pulse-Session-Id header |
| Audit-sink-failure behaviour decided with Nematullah and implemented | Proposed, pending ratification - see O2 |
| No-write CI check live in the Lite repository | Stand-in only - see docs/scope.md |
| Secrets resolving from the managed store, TLS enforced on the listener | Secrets: done (env-var seam). TLS: infra concern, see docs/scope.md |
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。