Self-hosted MCP gateway for registering HTTP and stdio MCP servers once, exposing them through a unified MCP endpoint, and using tool groups, access-control, and upstream authentication for shared agent deployments.
MCPJungle centralizes access to many upstream MCP servers, so a broad gateway token can expose every registered tool, prompt, and resource., Registered stdio MCP servers run as child processes on the MCPJungle host or container and inherit the configured command, args, environment, and mounted filesystem access., Upstream HTTP servers can store bearer tokens, custom headers, and beta OAuth credentials in gateway configuration., Tool groups can reduce context and capability exposure, but they must be reviewed whenever upstream servers or tools change., Use enterprise mode, explicit client allow-lists, least-privilege upstream credentials, TLS, and audit/observability controls for shared deployments.
Privacy notes
MCPJungle can store registered server URLs, command arguments, environment variables, headers, bearer tokens, OAuth tokens, client access tokens, tool schemas, prompt names, resource URIs, and gateway database state., Tool calls, prompts, resources, logs, metrics, CLI output, dashboard views, and upstream responses may contain sensitive project data or third-party service data., The proxy source strips client-sent headers before forwarding upstream calls, but configured upstream credentials and tool payloads remain sensitive., Redact gateway URLs, access tokens, upstream tokens, OAuth client secrets, command args, environment values, resource URIs, and tool outputs before sharing logs or screenshots.
Author
mcpjungle
Submitted by
oktofeesh1
Claim status
unclaimed
Last verified
2026-06-06
Decision playbook
Review trust signals before you adopt
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
Compare context
Selected
0
Current score
63
Baseline
—
Delta
No baseline selected
No major trust-signal divergence detected in the current selection.
Source and provenance checks
Needs review
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
No reviewed flag detected in metadata.
Pending
Safety and privacy checks
Complete
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done
Package and install checks
Needs review
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending
Compare-driven decision checks
Needs review
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
5 safety and 4 privacy notes across 4 risk areas. Review closely: credentials & tokens, third-party handling.
4 areas
SafetyCredentials & tokensMCPJungle centralizes access to many upstream MCP servers, so a broad gateway token can expose every registered tool, prompt, and resource.
SafetyLocal filesRegistered stdio MCP servers run as child processes on the MCPJungle host or container and inherit the configured command, args, environment, and mounted filesystem access.
SafetyCredentials & tokensUpstream HTTP servers can store bearer tokens, custom headers, and beta OAuth credentials in gateway configuration.
SafetyGeneralTool groups can reduce context and capability exposure, but they must be reviewed whenever upstream servers or tools change.
SafetyCredentials & tokensUse enterprise mode, explicit client allow-lists, least-privilege upstream credentials, TLS, and audit/observability controls for shared deployments.
PrivacyCredentials & tokensMCPJungle can store registered server URLs, command arguments, environment variables, headers, bearer tokens, OAuth tokens, client access tokens, tool schemas, prompt names, resource URIs, and gateway database state.
PrivacyThird-party handlingTool calls, prompts, resources, logs, metrics, CLI output, dashboard views, and upstream responses may contain sensitive project data or third-party service data.
PrivacyCredentials & tokensThe proxy source strips client-sent headers before forwarding upstream calls, but configured upstream credentials and tool payloads remain sensitive.
PrivacyCredentials & tokensRedact gateway URLs, access tokens, upstream tokens, OAuth client secrets, command args, environment values, resource URIs, and tool outputs before sharing logs or screenshots.
Disclosure: MPL-2.0-licensed open-source MCP gateway. It is a management and proxy layer, not a replacement for reviewing the safety of every upstream MCP server it exposes.
Safety notes
MCPJungle centralizes access to many upstream MCP servers, so a broad gateway token can expose every registered tool, prompt, and resource.
Registered stdio MCP servers run as child processes on the MCPJungle host or container and inherit the configured command, args, environment, and mounted filesystem access.
Upstream HTTP servers can store bearer tokens, custom headers, and beta OAuth credentials in gateway configuration.
Tool groups can reduce context and capability exposure, but they must be reviewed whenever upstream servers or tools change.
Use enterprise mode, explicit client allow-lists, least-privilege upstream credentials, TLS, and audit/observability controls for shared deployments.
Privacy notes
MCPJungle can store registered server URLs, command arguments, environment variables, headers, bearer tokens, OAuth tokens, client access tokens, tool schemas, prompt names, resource URIs, and gateway database state.
Tool calls, prompts, resources, logs, metrics, CLI output, dashboard views, and upstream responses may contain sensitive project data or third-party service data.
The proxy source strips client-sent headers before forwarding upstream calls, but configured upstream credentials and tool payloads remain sensitive.
Redact gateway URLs, access tokens, upstream tokens, OAuth client secrets, command args, environment values, resource URIs, and tool outputs before sharing logs or screenshots.
Prerequisites
Docker Compose or installed MCPJungle binary.
SQLite or PostgreSQL persistence selected for gateway state.
Upstream MCP server list, auth strategy, and transport type reviewed.
MCP client support for streamable HTTP, or a reviewed bridge for clients that need one.
Access-control plan for development mode, enterprise mode, tool groups, and client tokens.
Schema details
Install type
cli
Troubleshooting
No
Source repository stats
Scope
Source repo
Collection metadata
Estimated setup
20 minutes
Difficulty
advanced
Tool listing metadata
Disclosure
MPL-2.0-licensed open-source MCP gateway. It is a management and proxy layer, not a replacement for reviewing the safety of every upstream MCP server it exposes.
MCPJungle is a self-hosted MCP gateway for registering upstream MCP servers and
exposing them through a single MCP endpoint. It discovers tools, prompts, and
resources from registered servers, rewrites names into canonical
server__tool and server__prompt formats, and proxies client calls to the
correct upstream server.
Use it when an individual or team has more MCP servers than they want to wire
into every client separately, or when a shared deployment needs tool groups,
client allow-lists, upstream authentication, and centralized observability.
These sources were reviewed on 2026-06-06. Prefer the live repository,
quickstart, license, core concepts, registration guides, tool-group guide,
access-control guide, upstream-authentication guide, and MCP proxy source for
current setup and behavior.
Features
Expose a unified streamable HTTP MCP gateway endpoint.
Register upstream streamable HTTP MCP servers.
Register stdio MCP servers from JSON configuration files.
Discover and proxy upstream tools, prompts, and resources.
Use canonical names to avoid collisions across registered servers.
Create tool groups with curated tools or entire upstream servers.
Run development mode for personal/local use.
Run enterprise mode with MCP client identities and server allow-lists.
Store upstream bearer tokens, custom headers, and beta OAuth credentials for
registered HTTP MCP servers.
Use stateless sessions by default or stateful sessions for slow-starting
upstream servers.
Installation
Run MCPJungle with Docker Compose or the standalone binary, register upstream MCP
servers, and expose the gateway behind TLS before sharing it with agents:
For local development deployments that use a plain local endpoint, follow the
upstream quickstart and keep that configuration out of shared or production
clients.
Use Cases
Register Context7, filesystem, time, GitHub, and internal MCP servers once.
Give Claude one MCP endpoint instead of many individual server configs.
Expose only a curated set of tools through a tool-group endpoint.
Create separate client identities for Cursor, Claude, Copilot, and internal
agents.
Manage upstream bearer tokens or OAuth for hosted MCP servers in one gateway.
Avoid name collisions by using canonical server__tool tool names.
Run a shared team MCP gateway with PostgreSQL, observability, and access
controls.
Safety and Privacy
MCPJungle concentrates capability. Treat access to the gateway as access to the
registered upstream tools, prompts, resources, and credentials. In shared
deployments, use enterprise mode, create one client identity per integration,
avoid wildcard allow-lists, and prefer tool groups that expose only the tools a
client needs.
Be careful with stdio servers because they run as child processes on the
MCPJungle host or container and may receive environment variables or mounted
filesystem access. Upstream HTTP registrations can store bearer tokens, custom
headers, or OAuth credentials. Protect the gateway database, CLI config files,
dashboard, logs, metrics, resource URIs, tool payloads, and client access tokens.
Duplicate Check
No mcpjungle/MCPJungle, MCPJungle gateway, MCPJungle proxy, or matching source
URL entry was found in content/mcp or README.md. The existing MCPHub Server
Manager entry covers a different server-management project; it does not cover
MCPJungle's gateway, canonical proxying, tool groups, enterprise client
allow-lists, and upstream-authentication model.
Show that MCPJungle Gateway is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.
[](https://heyclau.de/entry/mcp/mcpjungle-gateway)
How it compares
MCPJungle Gateway side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
Self-hosted MCP gateway for registering HTTP and stdio MCP servers once, exposing them through a unified MCP endpoint, and using tool groups, access-control, and upstream authentication for shared agent deployments.
PortSwigger's Burp Suite MCP Server extension connects Burp Suite to MCP clients through an SSE server or packaged stdio proxy for request, Repeater, Intruder, history, scanner, Collaborator, and configuration workflows.
Self-hostable calendar MCP service for Keeper.sh that lets Claude list calendars and accounts, read events, create or update events, delete events, respond to invites, and retrieve iCal feed URLs through Keeper's API.
✓MCPJungle centralizes access to many upstream MCP servers, so a broad gateway token can expose every registered tool, prompt, and resource.
Registered stdio MCP servers run as child processes on the MCPJungle host or container and inherit the configured command, args, environment, and mounted filesystem access.
Upstream HTTP servers can store bearer tokens, custom headers, and beta OAuth credentials in gateway configuration.
Tool groups can reduce context and capability exposure, but they must be reviewed whenever upstream servers or tools change.
Use enterprise mode, explicit client allow-lists, least-privilege upstream credentials, TLS, and audit/observability controls for shared deployments.
✓Burp Suite MCP Server can send HTTP/1.1 and HTTP/2 requests, create Repeater tabs, send requests to Intruder, toggle Proxy Intercept, pause or resume Burp's task execution engine, and update the active message editor.
In Burp Suite Professional it can also expose scanner issues and generate or poll Collaborator payloads for out-of-band testing.
The extension includes approval flows for outbound HTTP requests and sensitive data access, but users can configure always-allow targets and disable some approval requirements.
Configuration editing tools can import project-level or user-level Burp options when enabled in the extension, which can change proxy, scanner, target, and other Burp behavior.
Use only on systems and applications where testing is authorized; active requests, Intruder traffic, scanner workflows, and Collaborator payloads can affect third-party services.
Keep the MCP server bound to trusted local interfaces and avoid exposing the SSE server to untrusted networks.
✓Coolify MCP Server can start, stop, restart, redeploy, cancel deployments, update env vars, create or delete projects, environments, applications, databases, services, backups, storages, scheduled tasks, private keys, GitHub apps, and cloud tokens depending on API permissions.
Batch tools such as `restart_project_apps`, `bulk_env_update`, `stop_all_apps`, and `redeploy_project` can affect multiple production services at once.
Deployment, control, backup, storage, private key, cloud token, GitHub app, and scheduled-task operations should require explicit confirmation.
Custom `--header` values may carry auth-proxy secrets; never let an agent invent, log, or modify them casually.
Test on non-production projects or staging resources before allowing Claude to operate live Coolify infrastructure.
✓Keeper's MCP tools can read calendar data and perform write operations including event creation, event updates, event deletion, and RSVP changes.
The MCP handler requires an authenticated bearer token and checks for the Keeper MCP read scope before exposing tools.
Tool calls proxy to Keeper API endpoints with the caller's bearer token, so the MCP client inherits the user's connected calendar permissions.
Start by listing calendars and reading bounded date ranges; require human confirmation before modifying or deleting events.
Review Google, Microsoft, iCloud, CalDAV, ICS, and self-hosted calendar account permissions before exposing them to an agent.
Privacy notes
✓MCPJungle can store registered server URLs, command arguments, environment variables, headers, bearer tokens, OAuth tokens, client access tokens, tool schemas, prompt names, resource URIs, and gateway database state.
Tool calls, prompts, resources, logs, metrics, CLI output, dashboard views, and upstream responses may contain sensitive project data or third-party service data.
The proxy source strips client-sent headers before forwarding upstream calls, but configured upstream credentials and tool payloads remain sensitive.
Redact gateway URLs, access tokens, upstream tokens, OAuth client secrets, command args, environment values, resource URIs, and tool outputs before sharing logs or screenshots.
✓Proxy HTTP history, WebSocket history, Organizer items, scanner issues, request and response bodies, headers, cookies, tokens, session identifiers, and Collaborator interaction data may be returned to the MCP client.
The extension can read project-level and user-level Burp configuration; upstream code filters some configuration credentials when configured, but users should still treat exported options as sensitive.
MCP prompts, responses, Burp logs, and client transcripts can retain target URLs, credentials, payloads, vulnerability details, and proprietary application behavior.
The stdio proxy and SSE server bridge Burp traffic into the MCP client process; keep client configs, proxy paths, and Burp project files protected.
✓Coolify access tokens, base URLs, custom headers, application UUIDs, server IPs, domains, logs, env vars, deployment logs, private keys, cloud-provider tokens, GitHub app data, team membership, backups, database metadata, and service configuration can be exposed to the MCP client.
Application and deployment logs may contain secrets, customer data, build output, container metadata, private repository details, and runtime errors.
Environment variable and cloud-token tools can reveal or mutate sensitive infrastructure credentials.
Documentation search is local to the server, but Coolify API calls contact the configured Coolify instance.
Keep tokens and auth-proxy headers in local MCP client configuration only, and avoid sharing transcripts that include infrastructure identifiers or logs.
✓Calendar events can expose titles, descriptions, locations, free/busy status, account names, provider names, calendar names, calendar URLs, and iCal feed URLs.
OAuth tokens, CalDAV credentials, session tokens, bearer tokens, auth secrets, encryption keys, database URLs, and MCP URLs are sensitive deployment data.
Keeper stores and syncs calendar data across connected providers; logs, MCP transcripts, API responses, and monitoring traces can include private schedule details.
Redact event ids, calendar ids, account names, feed URLs, bearer tokens, event text, locations, and provider details before sharing debug output.
Prerequisites
Docker Compose or installed MCPJungle binary.
SQLite or PostgreSQL persistence selected for gateway state.
Upstream MCP server list, auth strategy, and transport type reviewed.
MCP client support for streamable HTTP, or a reviewed bridge for clients that need one.
Burp Suite Community or Professional with Java extension support.
Java and the `jar` command available for building and loading the extension.
Gradle wrapper execution allowed for building `build/libs/burp-mcp-all.jar` from source.
An MCP client that can connect to the Burp SSE server or run the packaged stdio proxy.
Running Coolify instance with API access enabled.
Coolify API access token scoped to the resources Claude may inspect or manage.
Node.js 20 or newer for the published npm server.
Review of which servers, projects, applications, databases, services, deployments, env vars, private keys, teams, cloud tokens, and scheduled tasks the token can access.
Self-hosted or hosted Keeper.sh deployment with the MCP service available.
Keeper API, database, auth secret, and MCP public URL configured.
Authorized Keeper bearer token with the required MCP read scope.
Connected calendar accounts and calendars reviewed for read and write access.
Install
Run MCPJungle with Docker Compose or the `mcpjungle` binary, register upstream MCP servers, and point Claude at the gateway `/mcp` endpoint.
./gradlew embedProxyJar
npx -y @masonator/coolify-mcp
Deploy Keeper.sh with the MCP service enabled, configure `MCP_PUBLIC_URL`, and connect an MCP client to the Keeper MCP endpoint with an authorized bearer token.