Atlas / MCP servers / agentic-community / Gateway Registry

Gateway RegistryBLOCK

mcp/agentic-community/gateway-registry

Enterprise-ready MCP Gateway & Registry that centralizes AI development tools with secure OAuth authentication, dynamic tool discovery, and unified access for both autonomous AI agents and AI coding assistants. Transform scattered MCP server chaos into governed, auditable tool access with Keycloak/E

Verdict
BLOCK
Grade
F
Trust score
19 /100
Exposed tools
9 7r · 2w · 0d
Transport
stdio · streamable-http
License
Apache-2.0
Stars
944
01

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

Unified Agent & MCP Server Registry – Gateway for AI Development Tools

[](https://github.com/agentic-community/mcp-gateway-registry/stargazers) [](https://github.com/agentic-community/mcp-gateway-registry/network) [](LICENSE) [](https://github.com/agentic-community/mcp-gateway-registry/releases)

Get Running Now | Docs | Executive Brief | Slide Deck | Demo Videos | AWS Workshop | Community

The MCP Gateway & Registry is a single, governed control plane for every AI asset in your organization, from MCP servers and AI agents to skills and any custom asset your teams build. It is open source, licensed under Apache 2.0, and runs on Kubernetes (Amazon EKS), fully managed serverless (Amazon ECS), or Docker Compose (Amazon EC2).

It began as a gateway and registry for the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introdu

Read from source at commit d5525b8f8615OBSERVED · 2026-09-27
02

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code
claude mcp add mcp-gateway-registry-infra --env A2A_SCANNER_LLM_API_KEY=${A2A_SCANNER_LLM_API_KEY} --env AGENT_BEARER_TOKEN=${AGENT_BEARER_TOKEN} --env AGENT_TOKEN=${AGENT_TOKEN} --env AI_DEFENSE_API_KEY=${AI_DEFENSE_API_KEY} -- npx -y [email protected]
claude-desktop
{
  "mcpServers": {
    "mcp-gateway-registry-infra": {
      "command": "npx",
      "args": [
        "-y",
        "[email protected]"
      ],
      "env": {
        "A2A_SCANNER_LLM_API_KEY": "${A2A_SCANNER_LLM_API_KEY}",
        "AGENT_BEARER_TOKEN": "${AGENT_BEARER_TOKEN}",
        "AGENT_TOKEN": "${AGENT_TOKEN}",
        "AI_DEFENSE_API_KEY": "${AI_DEFENSE_API_KEY}"
      }
    }
  }
}
03

Exposed tools (9)

7 read · 2 write · 0 destructive.

ToolRiskDescription
GadgetreadA custom type
adminsreadAdmin group
current_time_by_timezoneread
currenttime-usersreadUsers with access to currenttime server
ds1read
mcp_commandreadCall MCP gateway commands (ping, list, call, init).
read_docsreadSearch and read documentation files from the docs folder. Use this when users ask questions about the project, features, setup, configuration, or troubleshooting.
registry_taskwriteRun service management, imports, user management, or diagnostics tasks.
shell_commandwriteRun a read-only diagnostic command. Only a fixed allowlist of inspection binaries
04

Trust audit

BLOCKgrade F · trust 19/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

LayerWhat it checksResult
L0Provenance & inventoryWARN
L1Static analysis of the codeFAIL
L2Instruction surface (what it tells the agent)FAIL
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
declared (16 observation(s))
Network
declared (15 observation(s))
Shell
declared (2 observation(s))
Dependencies
not all pinned
Secrets in source
found

Findings (25)

CRITICALHard-coded secrets · secret.private_key · CWE-798, CWE-321
.env.example:869
# GITHUB_APP_PRIVATE_KEY="-----BEGIN RSA PRIVATE KEY-----\n...\n-----END RSA PRIVATE KEY-----"
CRITICALHard-coded secrets · secret.private_key · CWE-798, CWE-321
terraform/aws-ecs/terraform.tfvars.example:1296
# github_app_private_key     = "-----BEGIN RSA PRIVATE KEY-----\\n...\\n-----END RSA PRIVATE KEY-----"
CRITICALPrompt injection · prompt.transfer_instruction · CWE-94, CWE-1427
.claude/skills/agentcore-register/SKILL.md:19
Send the MCP `initialize` request and capture the response headers (for the session ID) and body (for server info).
Why it matters. an instruction to move sensitive data to an outside destination
Fix. remove; a skill never needs the user's secrets off the machine
CRITICALPrompt injection · prompt.transfer_instruction · CWE-94, CWE-1427
.claude/skills/terraform-setup/SKILL.md:595
**Authentication:** with `REGISTRY_URL` and `KEYCLOAK_URL` exported, `register` fetches the JWT it needs dynamically (this is the same flow the README's post-deployment registration uses). **Do not re
Why it matters. an instruction to move sensitive data to an outside destination
Fix. remove; a skill never needs the user's secrets off the machine
CRITICALPrompt injection · prompt.transfer_instruction · CWE-94, CWE-1427
.claude/skills/terraform-setup/SKILL.md:758
After login the browser lands on `https://<cloudfront>/login?error=oauth2_callback_failed`. The auth-server logs (`/ecs/<name>-auth-server`) show its server-side token exchange returning **401** on `P
Why it matters. an instruction to move sensitive data to an outside destination
Fix. remove; a skill never needs the user's secrets off the machine
HIGHCode injection · code.deserialize · CWE-78, CWE-94, CWE-95
infra/lib/registry/registry-config.ts:706
parsed = yaml.load(raw) as Record<string, any> | null;
Why it matters. deserialises untrusted bytes into live objects
Fix. use json or yaml.safe_load
HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59
.claude/skills/usage-report/run_report.sh:36
SSH_KEY="${SSH_KEY:-$HOME/.ssh/id_ed25519}"
Why it matters. touches a credential store
HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59
docker-compose.yml:558
# AWS credentials are OPT-IN. Mounting the whole ~/.aws directory into an
Why it matters. touches a credential store
HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59
terraform/telemetry-collector/bastion-scripts/setup-bastion.sh:9
SSH_KEY="${SSH_KEY:-~/.ssh/id_ed25519}"
Why it matters. touches a credential store
HIGHNetwork egress · net.metadata_ip · CWE-200, CWE-319
.claude/skills/macos-setup/SKILL.md:865
The cloud metadata address (169.254.169.254) is never permitted by either
Why it matters. cloud metadata endpoint: the classic SSRF credential grab
HIGHNetwork egress · net.metadata_ip · CWE-200, CWE-319
.env.example:1381
# unless you allowlist them here. The cloud metadata address (169.254.169.254)
Why it matters. cloud metadata endpoint: the classic SSRF credential grab
HIGHNetwork egress · net.metadata_ip · CWE-200, CWE-319
.env.example:1830
# workload-credential endpoints: EC2 IMDS (169.254.169.254, fd00:ec2::254),
Why it matters. cloud metadata endpoint: the classic SSRF credential grab
HIGHNetwork egress · net.tls_off · CWE-200, CWE-319
auth_server/server.py:9234
"GATEWAY_GENERIC_TLS_VERIFY=false — the generic hop will NOT verify "
Why it matters. certificate verification is disabled
Fix. leave verification on
HIGHNetwork egress · net.tls_off · CWE-200, CWE-319
credentials-provider/generate_creds.sh:92
./generate_creds.sh -a keycloak -k https://kc.example.com
Why it matters. certificate verification is disabled
Fix. leave verification on
HIGHHard-coded secrets · secret.db_uri · CWE-798, CWE-321
.env.example:1516
#   MONGODB_CONNECTION_STRING=mongodb://admin:admin@mongodb:27017/mcp_registry?authMechanism=SCRAM-SHA-256&authSource=admin&retryWrites=false
HIGHPrompt injection · prompt.credential_read · CWE-94, CWE-1427
.claude/skills/macos-setup/SKILL.md:332
# Read from .env by init-keycloak.sh in Phase 10; see the generation block above.
Why it matters. asks the agent to read credentials
HIGHPrompt injection · prompt.credential_read · CWE-94, CWE-1427
.claude/skills/macos-setup/SKILL.md:610
content = open('.env').read()
Why it matters. asks the agent to read credentials
HIGHPrompt injection · prompt.credential_read · CWE-94, CWE-1427
.claude/skills/macos-setup/SKILL.md:620
open('.env', 'w').write(content)
Why it matters. asks the agent to read credentials
HIGHPrompt injection · prompt.credential_read · CWE-94, CWE-1427
.claude/skills/macos-setup/SKILL.md:838
content = open('.env').read()
Why it matters. asks the agent to read credentials
HIGHPrompt injection · prompt.credential_read · CWE-94, CWE-1427
.claude/skills/macos-setup/SKILL.md:842
open('.env', 'w').write(content)
Why it matters. asks the agent to read credentials
MEDIUMInventory / provenance · inv.binary · CWE-1104
metrics-service/tmp/metrics.db
metrics.db
Why it matters. a compiled or binary member cannot be reviewed from source
Fix. ship source, or explain the binary in the README
MEDIUMInventory / provenance · inv.binary · CWE-1104
metrics-service/tmp/test.db
test.db
Why it matters. a compiled or binary member cannot be reviewed from source
Fix. ship source, or explain the binary in the README
MEDIUMAuth / authz · auth.debug_bypass · CWE-287, CWE-862
charts/keycloak-configure/templates/configmap.yaml:1068
# Check if admin password is set
MEDIUMAuth / authz · auth.debug_bypass · CWE-287, CWE-862
keycloak/setup/clean-keycloak.sh:68
# Check if admin password is set
MEDIUMAuth / authz · auth.debug_bypass · CWE-287, CWE-862
keycloak/setup/get-all-client-credentials.sh:36
# Check if admin password is set

Gates applied: critical_finding, no_behavioural_pass, undeclared_transfer.

Audited 2026-09-27 · audit v0.4.1 · source sha d5525b8f8615full audit observations/trust-audit/mcp-server/agentic-community__gateway-registry.json · Report an issue / request a re-scan
05

Audit history

Every audit this server has had. A grade with a past is a grade somebody is still checking.

DateSourceVerdictGradeScoreChange
2026-09-27d5525b8f8615BLOCKF19first audit
06

Questions

What is the Gateway Registry MCP server?

Enterprise-ready MCP Gateway & Registry that centralizes AI development tools with secure OAuth authentication, dynamic tool discovery, and unified access for both autonomous AI agents and AI coding assistants. Transform scattered MCP server chaos into governed, auditable tool access with Keycloak/E

What tools does Gateway Registry expose?

9 in total: 7 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.

Is Gateway Registry safe to connect to an agent?

No — not without reading the findings first. The audit graded it F (19/100) and found 20 critical or high issues in the source. Each one is listed on this page with the file and line it is on.

What credentials does Gateway Registry need?

It reads A2A_SCANNER_LLM_API_KEY, AGENT_BEARER_TOKEN, AGENT_TOKEN, AI_DEFENSE_API_KEY, ANS_API_KEY, ANS_API_SECRET, ANTHROPIC_API_KEY, API_KEY_HASH_ALGORITHM, ASOR_ACCESS_TOKEN, ASOR_CLIENT_SECRET, AUTH0_AUDIENCE and AUTH0_CLIENT_ID from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.

How does Gateway Registry run?

It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on npm as mcp-gateway-registry-infra at 1.0.0.

How current is this page?

The grade is for one exact copy of the source (d5525b8f8615), read on 2026-09-27. The repository is watched and re-audited when it changes.

Advertisement