DEV.co
MCP Servers · ascending-llc

jarvis-registry

Jarvis Registry is an open-source gateway that connects AI copilots and autonomous agents to enterprise tools through a single, secure entry point. It handles identity, access control, and full audit trails so teams can safely deploy AI agents without fragmented integrations or security gaps.

Source: GitHub — github.com/ascending-llc/jarvis-registry
2k
GitHub stars
300
Forks
Python
Primary language
Apache-2.0
License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

FieldValue
Repositoryascending-llc/jarvis-registry
Ownerascending-llc
Primary languagePython
LicenseApache-2.0 — OSI-approved
Stars2k
Forks300
Open issues1
Latest releaseasc0.5.1 (2026-07-06)
Last updated2026-07-07
Sourcehttps://github.com/ascending-llc/jarvis-registry

What jarvis-registry is

An Apache-2.0 licensed MCP (Model Context Protocol) gateway and A2A agent orchestration platform written in Python. Features OAuth 2.0/OIDC identity integration, fine-grained ACL enforcement, semantic skill discovery, OpenTelemetry tracing, and Prometheus metrics for enterprise governance of AI agent access.

Quickstart

Get the jarvis-registry source

Clone the repository and explore it locally.

terminalbash
git clone https://github.com/ascending-llc/jarvis-registry.gitcd jarvis-registry# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Secure AI Copilot Access to Internal Tools

Deploy Claude Desktop, Cursor, GitHub Copilot, or VS Code integrated with MCP to access enterprise databases, APIs, and services through one authenticated gateway with role-based controls.

Multi-Agent Orchestration with Governance

Coordinate autonomous worker agents through a central orchestrator with audit logging, semantic skill discovery, and per-agent permission boundaries to prevent unauthorized tool access.

Enterprise AI Adoption Without Security Fragmentation

Replace per-copilot or per-agent point integrations with a unified gateway that enforces identity (Keycloak, Cognito, Entra ID), ACLs, and observability across all AI clients in one platform.

Implementation considerations

  • Requires Docker Compose or Kubernetes (EKS, AKS, GKE) for production; Python virtual environment needed for local dev. Deployment complexity varies by cloud choice.
  • Mandatory identity provider setup (Keycloak, Cognito, or Entra ID) before agents can authenticate; misconfiguration will block agent access.
  • ACL and scope configuration is granular but manual; plan data access policies and role hierarchies before agents are registered.
  • OpenTelemetry and Prometheus integration assumed for production observability; requires collector/scraper infrastructure if not already present.
  • MCP server and A2A agent registration is declarative; ensure inventory and skill descriptions are maintained in sync with actual tool capabilities.

When to avoid it — and what to weigh

  • No Need for MCP Protocol — If your AI clients do not use or plan to adopt the Model Context Protocol (MCP), Jarvis Registry's core value proposition does not apply. Consider direct API authentication instead.
  • Lightweight, Single-Agent Use Cases — For simple, single-user or single-agent scenarios with minimal compliance needs, the operational overhead of running a full gateway, IDP, and observability stack may not justify the cost.
  • Existing Mature Agent Orchestration Layer — If you already have a proprietary or third-party agent orchestration, workflow, and governance platform in place, integrating Jarvis Registry as a gateway may create redundancy.
  • Strict Air-Gapped or Custom Auth Requirements — Jarvis Registry relies on OAuth 2.0/OIDC with external IDPs (Keycloak, Cognito, Entra ID). Air-gapped environments or highly custom identity systems may require significant customization.

License & commercial use

Licensed under Apache License 2.0 (Apache-2.0). This is a permissive open-source license that allows commercial use, modification, and distribution under stated conditions (retain license and copyright notices, state changes, provide license copy, disclaim warranties).

Apache-2.0 explicitly permits commercial use without royalty or proprietary obligation. You may use, modify, and redistribute Jarvis Registry in commercial products. No vendor lock-in to ASCENDING Inc.; however, no formal commercial support, SLA, or indemnification is provided by the license itself. Enterprise support is not mentioned in the provided data; contact ASCENDING Inc. separately.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

SignalAssessment
MaintenanceActive
DocumentationStrong
License clarityClear
Deployment complexityModerate
DEV.co fitGood
Assessment confidenceHigh
Security considerations

Platform is designed with identity and access control as core: OAuth 2.0/OIDC, fine-grained ACL, audit logging, and OpenTelemetry tracing. Security design documentation is available. No claims of third-party penetration testing or formal security certification in provided data. Relies on external IDPs (Keycloak, Cognito, Entra ID) for identity assurance; their security posture is inherited. Audit logging and observability enable post-incident forensics. For sensitive deployments, review security design docs and conduct threat modeling; engage ASCENDING Inc. for penetration testing or compliance certification if required.

Alternatives to consider

Anthropic's Model Context Protocol (direct)

Adopt MCP directly without a gateway; reduces overhead but requires custom auth, ACL, and observability at the application layer. Suitable only if you control all AI clients and agents.

LangChain Agents + Custom Gateway

Build a custom orchestration and gateway layer using LangChain; offers full control but requires significant engineering and does not include built-in identity, ACL, or enterprise observability.

Anthropic Workbench or Third-Party Agent Platforms

Use SaaS agent orchestration (e.g., Anthropic Workbench, Retool, n8n); trades operational overhead for managed infrastructure, but may lock you into a vendor and limit MCP extensibility.

Software development agency

Build on jarvis-registry with DEV.co software developers

Evaluate Jarvis Registry for your enterprise. Review the documentation at jarvisregistry.com, test the Quick Start locally, and contact ASCENDING Inc. for production support and custom integrations.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

Related on DEV.co

Explore the category and the services that help you build with it.

jarvis-registry FAQ

Does Jarvis Registry require my AI copilots to be MCP-compatible?
Yes. Jarvis Registry is fundamentally an MCP gateway. All AI copilots and agents must speak the Model Context Protocol (over SSE or Streamable HTTP) to communicate with it. Claude Desktop, Cursor, GitHub Copilot, and VS Code with MCP support are compatible.
Can I use Jarvis Registry without an external identity provider?
No. The platform is designed around OAuth 2.0/OIDC integration with external IDPs (Keycloak, Amazon Cognito, Microsoft Entra ID). Local or custom auth is not mentioned in the provided data; review the documentation or contact ASCENDING Inc. for alternatives.
Is Jarvis Registry suitable for small teams or startups?
Possibly, but with caveats. For a single developer or very small team, the operational overhead (Docker/Kubernetes, IDP setup, observability) may outweigh the benefit. Best fit is teams with 5+ engineers deploying multiple AI agents or copilots in a regulated industry.
What is the difference between MCP gateway and A2A orchestration?
MCP gateway secures AI copilots (human-facing, like Claude Desktop) accessing enterprise tools. A2A (agent-to-agent) orchestration coordinates autonomous agents (agent-to-agent communication) through the same secure gateway. Jarvis Registry does both.

Software developers & web developers for hire

DEV.co helps companies turn open-source tools like jarvis-registry into production software. Our software development services cover the full lifecycle — architecture, web development, integration, and maintenance — delivered by software developers and web developers who ship. Engage our software development agency to implement or customize it for your mcp servers stack.

Ready to Deploy Secure AI Agents?

Evaluate Jarvis Registry for your enterprise. Review the documentation at jarvisregistry.com, test the Quick Start locally, and contact ASCENDING Inc. for production support and custom integrations.