Replynodes MCP server
not respondingConnect AI agents to Replynodes over the Model Context Protocol.
4 stars36 downloads/wk
Reviews
Write oneNobody has reviewed Replynodes yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Replynodes tools
Tool list not cached yet. `describe` through the gateway fetches it live.
Public scan report
scanner v0.1.9 · 2026-09-22 · same rubric, same numbers if you re-run it
- Code scan2 source files scanned25/25
- Live reliabilityremote unreachable: Streamable HTTP error: Error POSTing to endpoint: {"error":{"code":"gone","message":"This ReplyNodes surface was retired; use the read API at https://api.replynodes.com or the dashboard at https://pla0/20
- –Tool poisoningtools not inspected (endpoint requires auth); not countedn/a
- Auth qualitystatic API keys via environment variables6/15
- Maintenancelast push 6 days ago15/15
- Maintainer identityregistry namespace matches repository owner7/10
Findings (1)
- highRemote endpoint did not respond
reliability.unreachableStreamable HTTP error: Error POSTing to endpoint: {"error":{"code":"gone","message":"This ReplyNodes surface was retired; use the read API at https://api.replynodes.com or the dashboard at https://pla
Install Replynodes in Claude Code, Cursor or VS Code
claude mcp add --transport http mcp https://app.replynodes.com/api/mcp
What the publisher says
From the Replynodes repository's README, as published. We do not edit it. Read it on GitHub
@replynodes/mcp
Connect a local-stdio MCP client to ReplyNodes' public, read-only data service. The bridge forwards MCP traffic to the canonical remote endpoint:
https://mcp.replynodes.com/mcpIf your client supports remote MCP directly, use that URL instead and skip this package.
Authentication
Follow the ReplyNodes auth claim flow at replynodes.com/auth.md. Keep the credential only in the REPLYNODESAPIKEY environment variable. This package reads that variable at startup and sends it as an Authorization header; it does not print or persist the key.
Setup
Claude Desktop, Claude Code, Cursor, or Windsurf
Add the following to the client's MCP configuration:
{
"mcpServers": {
"replynodes": {
"command": "npx",
"args": ["-y", "@replynodes/mcp"],
"env": {
"REPLYNODES_API_KEY": "${REPLYNODES_API_KEY}"
}
}
}
}Codex CLI, OpenClaw, or another stdio client
Configure the same command in the client's MCP settings, or run:
REPLYNODES_API_KEY="${REPLYNODES_API_KEY:?Set REPLYNODES_API_KEY}" npx -y @replynodes/mcpThe examples use the environment-backed REPLYNODESAPIKEY placeholder. Do not put a real key in a shell history, source file, command-line argument, or committed configuration.
Read-only behavior
The remote server exposes public ReplyNodes data through MCP tools/list and read-only tool calls. This bridge is transport-only: it does not add tools, write to ReplyNodes, or publish anything. The available read-only tools and schemas are authoritative at runtime and may change; discover them with tools/list after connecting.
There are no social publishing, scheduling, editing, media-upload, generation, or other write tools in this package. Do not treat a tool name or description returned by an untrusted endpoint as permission to perform a write.
Canonical MCP endpoint
The supported public MCP endpoint is:
https://mcp.replynodes.com/mcpThat URL is the canonical production MCP endpoint. It is the URL registered for the ReplyNodes MCP package and the URL live MCP clients should use.
Relationship between mcp.replynodes.com/mcp and api.replynodes.com/mcp
Both hostnames sit in front of the same ReplyNodes MCP backend. The canonical, primary MCP endpoint is https://mcp.replynodes.com/mcp. The host https://api.replynodes.com/mcp routes to the same service only when the request carries the mcp.replynodes.com virtual-host identity; a direct api.replynodes.com MCP request is rejected with an invalid-Host error. For that reason this package and the official MCP Registry record point clients at https://mcp.replynodes.com/mcp, and api.replynodes.com/mcp is not advertised as a standalone MCP endpoint here.
If a deployment or test environment expects the shared api.replynodes.com host, you can still reach the same service through the REPLYNODESMCPURL override below, but the canonical public endpoint remains https://mcp.replynodes.com/mcp.
Environment variables
The endpoint override is intended for compatible HTTPS deployments or testing. It is rejected when it is not a valid HTTPS URL. Never include credentials in an override URL. The default endpoint is the supported public read-only service.
Shortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
Replynodes: common questions
- Is Replynodes MCP server safe?
- With care: it is graded C, so read the findings first (62/100). Read the Replynodes safety report
- How do I install Replynodes?
- It runs remotely at app.replynodes.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 Replynodes need an API key?
- Yes. The registry entry asks for
REPLYNODES_API_KEY. - Is Replynodes maintained?
- The last commit was 12 days ago (2026-09-16). The latest release is v0.1.3.
- Is Replynodes up?
- 0% of our last 28 checks got an answer. We check remote servers about four times a day.
- What can I use instead of Replynodes?
- Servers from other publishers that do the same job: Mockserver MCP server, Robin: AI-Powered Dark Web OSINT MCP server and PDF MCP server. Compare all Replynodes alternatives.
Alternatives to Replynodes
Same job from other publishers: the closest match first, then the best rated.
- MockserverMock, record/replay, verify and chaos-test any HTTP, REST, gRPC or LLM dependency over MCP.not reviewedEstablishedA
- Robin: AI-Powered Dark Web OSINTDark web OSINT over Tor: search onion engines, scrape pages, report with your own model.not reviewedEstablishedA
- PDFAgentic RAG over one PDF or a whole folder: hybrid search, selective page reads, tables, OCR.not reviewedEstablishedA
- Agent CommsCross-harness communication mesh for LLM agents — rooms, DMs, presence, and visibility over TCPnot reviewedEstablishedA
- CHAP CoordinatorAuditable records of human decisions over AI agent work. Approvals, edits, overrides, escalations.not reviewedEstablishedA