Loppee MCP server
online · 100% uptimecom.loppee/loppee·v0.1.4Agent-first US business trust registry with neutral Trust Cards and published evidence.
0 stars
Reviews
Write oneNobody has reviewed Loppee yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Loppee tools (5)
write = sends, deletes, buys or postscompare_businessesFreeUse this when the user wants a side-by-side comparison of two to twenty known published Loppee business ids. Returns each business's exact class, disclosure, review context, and citation; unknown or unavailable ids are listed as unresolved rather than fabricated.
explain_recommendationFreeUse this when the user asks why a known published business is or is not recommendable on Loppee. Returns the published class, disclosure, recommendation basis, commercial-neutrality statement, and citation without exposing private weighting.
get_business_reviewsFreeUse this when the user wants the public customer reviews for one Loppee business id. Supports pagination and star-rating filters, preserves reviewer and interaction labels, masks reviewer identity, and changes nothing.
get_trust_cardFreeUse this when a business is already identified and the user wants its public Loppee Trust Card. Returns the exact business status, disclosure, coarse public city/state, public facts, review context, and citation without account authority, street address, ZIP, or signed media links.
lookup_businessFreeUse this when the user names a business and wants its Loppee Registry match. Returns exact and genuine-prefix name matches with stable ids, coarse public city/state, and truthful Verified, Listed, or Registry disclosure. It does not perform category discovery, accept or return caller location, or return a street address or ZIP.
Public scan report
scanner v0.1.10 · 2026-10-31 · same rubric, same numbers if you re-run it
- –Code scanremote-only server, no package to scann/a
- Live reliabilityremote reachable in 994ms20/20
- Tool poisoning5 tool descriptions checked15/15
- Auth qualityopen endpoint, read-only tools10/15
- Maintenancelast push 35 days ago12/15
- Maintainer identityendpoint served from the registry-verified namespace domain8/10
How Loppee compares
Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 87/100
- higher than 78% of 4,460 AI and agents servers
- Tools
- 5
- the typical AI and agents server has 6
- Tools that change things
- 0%
- typical: 0%. Fewer write tools means less an agent can break
- Uptime
- 100%
- of our recent checks; the typical remote AI and agents server answers 100%
- Response time
- 206 ms
- faster than 64% of 2,331 AI and agents servers
What its tools let an agent do: 5 read
Release history: 2 versions in the registry since 2026-08-24, latest 2026-09-26, typically one every 33 days.
Install Loppee in Claude Code, Cursor or VS Code
claude mcp add --transport http loppee https://loppee.com/mcp/discovery
What the publisher says
From the Loppee repository's README, as published. We do not edit it. Read it on GitHub
Loppee MCP
Public metadata for Loppee's narrow, no-auth MCP discovery surface.
Connect
- MCP URL: https://loppee.com/mcp/discovery
- Transport: Streamable HTTP
- Authentication: none
- Registry identity: com.loppee/loppee
- Source repository: https://github.com/loppee-ai/loppee-mcp
Configure the URL above as a remote HTTP MCP server in a compatible client. This repository does not contain credentials and does not require an API key.
Claude custom connector
Add Loppee to Claude prefills Claude's custom-connector dialog. Claude still asks the signed-in user to review and confirm the name and URL; the link does not publish a directory listing or grant permissions.
The canary and directory-review packet is in distribution/claude-connectors-directory.md.
Cursor
Copy the following official Cursor MCP install link into a browser with Cursor installed. Cursor shows the decoded remote-server configuration and asks the user to confirm before installation:
cursor://anysphere.cursor-deeplink/mcp/install?name=loppee&config=eyJ1cmwiOiJodHRwczovL2xvcHBlZS5jb20vbWNwL2Rpc2NvdmVyeSJ9The decoded configuration is exactly:
{
"url": "https://loppee.com/mcp/discovery"
}For manual or project setup, use mcp.json. The candidate Cursor Marketplace manifest is .cursor-plugin/plugin.json. Neither file contains an account, header, credential, key, or local command. The Marketplace preflight is in distribution/cursor-marketplace.md.
OpenClaw
The metadata-only Agent Plugins Bundle source is in clawhub/loppee-discovery. It contains no skill, runtime code, credential, header, key, customer data, owner tool, write tool, or location input.
The current ClawHub CLI incorrectly requires a native openclaw.plugin.json from compatible bundles. Loppee will preserve the narrower Agent Plugins Bundle instead of changing it into a native plugin or publishing it as a Skill. The live admission blocker and reproduction are tracked in openclaw/clawhub#3513. See distribution/openclaw-clawhub.md for the verified package boundary and publication gate.
Public tool boundary
The endpoint exposes exactly five read-only named-business research tools:
Every tool is declared read-only, non-destructive, and closed-world. This surface exposes no category discovery, address or location input, pricing, checkout, customer account, business-owner, write, or Back Office authority.
Privacy and neutrality
handoff, or device-location input.
- The tool schemas accept no address, city, state, ZIP, coordinates, location
or recommendation explanation on this surface.
- The endpoint returns no private customer or owner data.
- Payment does not change lookup, comparison, verification, reviews, evidence,
- Missing or unresolved records must remain unresolved rather than fabricated.
Loppee Personal, authenticated consumer MCP, Loppee Business, and Back Office are separate products with different authorization boundaries. They are not published through this repository.
Registry metadata
Shortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
Grade history
- 2026-10-31restoreC → Ascore 87
Loppee: common questions
- Is Loppee MCP server safe?
- Yes, by our scan: it is graded A (87/100). Read the Loppee safety report
- How do I install Loppee?
- It runs remotely at loppee.com. Add it to Claude Code, Cursor, VS Code or Claude Desktop with the snippets above, or call it through the mcp.market gateway without installing anything.
- Does Loppee need an API key?
- Not as far as the registry entry and our scan can tell: no credentials are declared or required.
- Is Loppee maintained?
- The last commit was 39 days ago (2026-09-26). The latest release is v0.1.4.
- Is Loppee up?
- 100% of our last 28 checks got an answer. We check remote servers about four times a day.
- What can I use instead of Loppee?
- Servers from other publishers that do the same job: HOL Guard MCP server, Basic Memory MCP server and Agent-Native Chat MCP server.
- How does Loppee compare with other AI and agents MCP servers?
- safety score 87/100 (higher than 78% of 4,460 AI and agents servers); tools 5 (the typical AI and agents server has 6); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.
Alternatives to Loppee
Same job from other publishers: the closest match first, then the best rated.
- HOL GuardLocal-first AI agent security evidence and approval workflows through HOL Guard's stdio MCP server.not reviewedEstablishedA
- Basic MemoryLocal-first knowledge management with bi-directional LLM sync via Markdown files.not reviewedWidely usedB
- Agent-Native ChatMinimal chat-first app with durable threads, actions, and the app-agent loopnot reviewedEstablishedA
- Watch SkillWatch video and live sessions, keep timestamped evidence, and verify an agent's own work.not reviewedEstablishedB
- OpenWorkProofVerifiable agent evidence; opt-in Codex guard and reference SOC alert review.not reviewedEstablishedA
