Server data from the Official MCP Registry
Agent-facing lodging decision tool. Non-sensitive categories only. Alpha, unauthenticated.
About
Agent-facing lodging decision tool. Non-sensitive categories only. Alpha, unauthenticated.
Remote endpoints: streamable-http: https://di-001-a-lodging-decision-mcp.scenicstay.workers.dev/mcp
Security Report
Valid MCP server (2 strong, 2 medium validity signals). No known CVEs in dependencies. Imported from the Official MCP Registry.
1 tool verified · Open access · No issues found
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
Permissions Required
This plugin requests these system permissions. Most are normal for its category.
How to Connect
Remote Plugin
No local installation needed. Your AI client connects to the remote endpoint directly.
Add this to your MCP configuration to connect:
{
"mcpServers": {
"io-github-scenicstay-lodging-decision-mcp": {
"url": "https://di-001-a-lodging-decision-mcp.scenicstay.workers.dev/mcp"
}
}
}Documentation
View on GitHubFrom the project's GitHub README.
Lodging Decision MCP Server (alpha)
Narrow, agent-facing MCP server wrapping PR #25's (Scenic-Stay/staygraph, branch
agent/scenicgraph-lodging-decision-api) lodging-decision kernel unmodified. Governed by
gauntlet/projects/decision-intelligence-v0.1/WORK-ORDERS.yaml
(DI-001-A) in Scenic-Stay/scenic-intelligence. Read that Work Order before changing scope.
Live: https://di-001-a-lodging-decision-mcp.scenicstay.workers.dev/mcp
Extracted from Scenic-Stay/staygraph (services/di-001-a-mcp-decision-server/, branch
agent/di-001-a-mcp-decision-server, commits 8c8f4c7/f7df7b3) into its own public repo so
GitHub-based MCP discovery (the Official MCP Registry, Glama's crawler) can find it without
exposing the rest of staygraph's code. That branch is preserved as historical record; deploy
from this repo going forward.
What this is
One MCP tool, lodging_decision. Callers supply a traveler profile, trip context, and
a list of candidate listings; the tool returns a deterministic, evidence-backed
recommendation with score breakdown, tradeoffs, risk flags, missing information, and
confidence. It does not search inventory, book, transact, or persist anything.
Scope
Non-sensitive categories only: budget, location, amenities, quality/reviews,
cancellation policy, fees, remote-work/family/business/relocation/event trip framing,
and stated accessibility needs. No Safety & Belonging, Medical Recovery, or
protected-characteristic-adjacent input is accepted — see src/guard.ts, a
defense-in-depth denylist that rejects (does not silently strip) free-text fields
referencing a protected characteristic or discriminatory steering criterion, on top of
the fact that the underlying kernel has no such fields to begin with.
Structure
src/kernel/— PR #25's contracts, scorer, validation, and fixtures, copied byte-for-byte (test/kernel-parity.test.tsproves this against PR #25's own six tests and committed example fixtures).src/guard.ts— the denylist guard described above.src/worker.ts— Cloudflare Worker entrypoint: MCPWebStandardStreamableHTTPServerTransportatPOST /mcp, plus a KV-backed soft daily request ceiling and a/healthcheck.test/— kernel parity tests, guard tests.
Local development
npm install
npm test # kernel parity + guard tests
npm run typecheck
npm run dev # wrangler dev, serves http://127.0.0.1:8787/mcp
Deploy
npm run deploy # wrangler deploy
Requires the DI_001_A_ABUSE_CEILING KV namespace bound in wrangler.jsonc (already
provisioned; see the Work Order's execution_result for the namespace ID and actual
Cloudflare cost recorded at deploy time).
Known limitations (carried over from PR #25, unresolved here)
Unauthenticated, unversioned, no rate limiting beyond the soft daily ceiling, not calibrated against real human booking decisions. Do not expose to untrusted high-volume traffic without a further hardening decision — see the Work Order's circuit breakers for the manual kill-switch (pull the Worker; delist from all directories) if abuse signals appear.
Reviews
No reviews yet
Be the first to review this server!
More Developer Tools MCP Servers
Fetch
Freeby Modelcontextprotocol · Developer Tools
Web content fetching and conversion for efficient LLM usage
Toleno
Freeby Toleno · Developer Tools
Toleno Network MCP Server — Manage your Toleno mining account with Claude AI using natural language.
mcp-creator-python
Freeby mcp-marketplace · Developer Tools
Create, build, and publish Python MCP servers to PyPI — conversationally.
MarkItDown
Freeby Microsoft · Content & Media
Convert files (PDF, Word, Excel, images, audio) to Markdown for LLM consumption
MCP Marketplace
Freeby mcp-marketplace · Developer Tools
Search and install MCP servers from inside your AI client.
FinAgent
Freeby mcp-marketplace · Finance
Free stock data and market news for any MCP-compatible AI assistant.
