storedge-mcp

storedge-mcp

MCP server for StorEdge self-storage management platform, enabling agents to manage tenants, units, leads, billing, insurance, tasks, and move-in/out pipelines via natural language.

Category
访问服务器

README

storedge-mcp

Tests Python License MCP

The first MCP server for StorEdge (Storable Edge) - the self-storage management platform that 5,000+ storage facilities rely on.

Read and write tenants, units, leads, ledgers, move-ins/outs, insurance, tasks, and surcharge calculations for a self-storage facility - straight from Claude, Cursor, or any MCP client.


Why this exists

StorEdge is the operations backbone for tens of thousands of self-storage facilities (under the Storable / Red Nova Labs umbrella). Its REST API at api.storedgefms.com powers tenant management, lead tracking, billing, insurance enrollment, task workflows, and move-in/out pipelines.

The existing integration options all require building OAuth 1.0 signing + paginated list handling + typed-error mapping from scratch every time:

  • Tray.ai connector - locked inside Tray's walled garden, no MCP, no local-first.
  • Twilio integration - single-purpose, voice-channel only.
  • Java/PHP/.NET sample clients (rednovalabs/storedge-api-client-*) - boilerplate code, no Claude tool surface.
  • No first-party MCP server - StorEdge has not shipped one.

storedge-mcp is the open MCP surface. One stdio call away, agents can answer "what 10x10 units are available right now?", "create a reservation for unit A-12", "show me who's past due this week", and "calculate the surcharge for a $125 Visa payment".


Quick start

pip install -e .
export STOREDGE_CONSUMER_KEY=...      # from StorEdge dashboard
export STOREDGE_CONSUMER_SECRET=...   # from StorEdge dashboard
export STOREDGE_FACILITY_ID=<uuid>    # optional; tools accept it as arg too
storedge-mcp

In Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "storedge": {
      "command": "storedge-mcp",
      "env": {
        "STOREDGE_CONSUMER_KEY": "...",
        "STOREDGE_CONSUMER_SECRET": "...",
        "STOREDGE_FACILITY_ID": "..."
      }
    }
  }
}

Tools (26)

Diagnostic

  • health_check(facility_id?) - verify credentials by listing one tenant. Returns fast-fail if OAuth signing is wrong.

Tenants

  • list_tenants(facility_id?, per_page?, page?, created_after?, created_before?, updated_since?, only_ids?)
  • list_current_tenants(facility_id?, per_page?, page?) - only currently-active tenants
  • search_tenants(account_code?, phone_number?, facility_id?) - by account code OR phone (exactly one)
  • get_tenant(tenant_id, facility_id?)
  • get_tenant_ledgers(tenant_id, facility_id?) - list of a tenant's rental agreements
  • get_tenant_notes(tenant_id, facility_id?) - operator notes attached to a tenant

Units

  • list_units(facility_id?, per_page?, page?, only_ids?) - every unit (rented + vacant + offline)
  • list_available_units(facility_id?, per_page?, page?) - only units available for move-in
  • get_unit(unit_id, facility_id?)
  • list_unit_groups(facility_id?, per_page?, page?) - size buckets (5x5, 10x10, etc.)
  • list_unit_types(facility_id?, per_page?, page?) - Parking, Climate, Drive-Up, etc.

Leads

  • list_leads(view="current", facility_id?, per_page?, page?) - view ∈ {current, reservations, inquiries, follow_up, waitlist, missed_move_ins}
  • get_lead_counts(facility_id?) - open lead counts by status
  • get_lead(lead_id, facility_id?)
  • create_lead(lead, facility_id?) - create a reservation or inquiry
  • update_lead(lead_id, lead, facility_id?) - PATCH an existing lead

Ledgers (rental agreements)

  • list_current_ledgers(facility_id?, per_page?, page?)
  • get_ledger(ledger_id, facility_id?)
  • get_ledger_delinquency(ledger_id, facility_id?) - overdue balance + days late
  • get_ledger_invoices(ledger_id, facility_id?)

Insurance

  • get_insurance_summary(facility_id?) - counts + revenue rollup
  • get_insurance_activity(facility_id?, per_page?, page?)

Tasks

  • list_tasks(category_id?, facility_id?, per_page?, page?)
  • list_task_categories(facility_id?)
  • create_task(task, facility_id?)
  • complete_task(task_id, facility_id?)

Move-in / move-out pipeline

  • list_move_ins(facility_id?, per_page?, page?) - in-progress move-ins
  • list_move_outs(facility_id?, per_page?, page?) - scheduled move-outs

Surcharge + invoices

  • calculate_surcharge(amount_cents, card_type?, facility_id?) - credit-card convenience fee quote
  • list_invoices_and_ledgers(facility_id?, per_page?, page?)

Auth

OAuth 1.0 one-legged (RFC 5849). Get your consumer key + secret from StorEdge dashboard → Corporate → Settings → API.

storedge-mcp ships its own OAuth 1.0 signer - not authlib's - because authlib's stock OAuth1Auth strips JSON request bodies during signing (silently breaking POST/PATCH/PUT). Our signer computes oauth_body_hash for non-form bodies and includes it in the signature per RFC 5849 §3.4.1.3.


Architecture

Built on industry-leading patterns from encode/httpx, stripe-python, lepture/authlib, boto3, pytest-dev/pytest, and astral-sh/ruff:

  • Shared httpx.AsyncClient with pooling + transport-level retries.
  • Typed exception hierarchy (StoredgeAuthError, StoredgeNotFoundError, StoredgeRateLimitError, StoredgeAPIError, StoredgeConnectionError) carrying structured error_code (from StorEdge's meta.error_code envelope) and request_id.
  • Application-level retry with exponential backoff + full jitter on 429/5xx, honoring Retry-After.
  • isError-compliance - every tool re-raises on failure so FastMCP sets isError: true on the wire (Blackwell audit / MCPTox).
  • JSONL audit logging via STOREDGE_AUDIT_LOG env var (or stderr default) - every tool call emits a structured record with redaction for password/api_key/token fields.
  • Property-based tests (Hypothesis) for JSON round-trips.
  • respx for httpx mocking - no live API calls in tests.

See engineering-playbook.md for the full pattern reference.


Development

python3 -m venv .venv
source .venv/bin/activate
pip install -e ".[dev]"

# Lint + type check + test
ruff check src tests
mypy src tests
pytest

# Coverage
pytest --cov=storedge_mcp --cov-report=term-missing

CI matrix (Python 3.10-3.13 on ubuntu/macos/windows) lives in .github/workflows/ci.yml.


Sister projects

sanjibani publishes MCP servers for one niche vertical SaaS per day. See the full list for the rest:

  • hawksoft-mcp - independent insurance agencies
  • open-dental-mcp - dental practice mgmt
  • ezyvet-mcp - veterinary
  • jobber-mcp - home service / field service
  • practicepanther-mcp - legal practice (solo/small)
  • realm-mp-mcp - church / nonprofit
  • fieldroutes-mcp - pest control / lawn care
  • cox-automotive-mcp - auto dealership
  • qualia-mcp - title and escrow
  • campspot-mcp - outdoor hospitality (campgrounds)
  • cleancloud-mcp - laundry / dry cleaning
  • playmetrics-mcp - youth sports clubs
  • foreup-mcp - golf course management

Need a custom MCP for your vertical SaaS stack? → mcp-services.landing.


License

MIT


References

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选