Borrowing it
Nothing to install: this file belongs to enkryptai/secure-mcp-gateway. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/enkryptai/secure-mcp-gateway/main/CLAUDE.mdgit clone --depth 1 https://github.com/enkryptai/secure-mcp-gatewayWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/enkryptai/secure-mcp-gateway/claude-md)<a href="https://agentmods.dev/instructions/enkryptai/secure-mcp-gateway/claude-md"><img src="https://agentmods.dev/badge/instructions/enkryptai/secure-mcp-gateway/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/enkryptai/secure-mcp-gateway/claude-md"><img src="https://agentmods.dev/badge/instructions/enkryptai/secure-mcp-gateway/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.13326 | $0.13326 |
| Opus 5.5 | $0.05330 | $0.05330 |
| Sonnet 5.5 | $0.02665 | $0.02665 |
| Haiku 4.5 | $0.01333 | $0.01333 |
Grade A, and why
secure-mcp-gateway CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 1,580 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Secure MCP Gateway - Complete Project Analysis
Version: 2.2.4 Last Updated: 2026-05-17 Project Type: Python Security Middleware for Model Context Protocol (MCP)
📋 Project Overview
Secure MCP Gateway is an Enkrypt AI-developed security middleware that sits between MCP clients (like Claude Desktop, Cursor) and MCP servers. It acts as both an MCP server (to clients) and an MCP client (to actual servers), providing:
- Authentication & Authorization
- OAuth 2.0/2.1 Support (Client credentials, mTLS, token management)
- Dynamic Tool Discovery
- Guardrails (Input/Output protection)
- Caching (Local & External Redis/KeyDB)
- Observability (OpenTelemetry, Prometheus, Grafana)
- RESTful API for management
🏗️ Project Structure
secure-mcp-gateway/
├── src/secure_mcp_gateway/
│ ├── __init__.py # Package initialization
│ ├── version.py # Version: "2.2.4"
│ ├── consts.py # Constants and defaults
│ ├── utils.py # Utilities, lazy logger, masking
│ ├── dependencies.py # Package dependencies list
│ │
│ ├── gateway.py # ⭐ Main MCP server (FastMCP)
│ ├── client.py # ⭐ MCP client to actual servers
│ ├── cli.py # ⭐ CLI interface (huge file)
│ ├── api_server.py # FastAPI REST API server (port 8001)
│ ├── api_routes.py # Additional API routes
│ ├── api_cache_routes.py # REST cache flush endpoints (port 8001)
│ ├── api_health_routes.py # REST health endpoints (port 8001)
│ ├── gateway_cache_routes.py # MCP gateway cache flush endpoints (port 8000)
│ │
│ ├── error_handling.py # Standardized error handling
│ ├── exceptions.py # Custom exception classes
│ │
│ ├── plugins/ # 🔌 Plugin system
│ │ ├── plugin_loader.py # Dynamic plugin discovery
│ │ ├── provider_loader.py # Provider loading utilities
│ │ │
│ │ ├── auth/ # Authentication plugins
│ │ │ ├── base.py # AuthProvider abstract class
│ │ │ ├── config_manager.py # Auth plugin manager
│ │ │ ├── enkrypt_provider.py # Enkrypt remote auth
│ │ │ └── example_providers.py # Local API key provider
│ │ │
│ │ ├── guardrails/ # Guardrail plugins
│ │ │ ├── base.py # GuardrailProvider abstract class
│ │ │ ├── config_manager.py # Guardrail plugin manager
│ │ │ ├── enkrypt_provider.py # Enkrypt guardrail API
│ │ │ └── example_providers.py # OpenAI, Custom keyword providers
│ │ │
│ │ └── telemetry/ # Telemetry plugins
│ │ ├── base.py # TelemetryProvider abstract class
│ │ ├── config_manager.py # Telemetry plugin manager
│ │ ├── opentelemetry_provider.py # OpenTelemetry OTLP
│ │ └── example_providers.py # Stdout provider
│ │
│ ├── services/ # 🎯 Service layer
│ │ ├── cache/
│ │ │ ├── cache_service.py # Core cache operations
│ │ │ ├── cache_management_service.py # Clear cache
│ │ │ └── cache_status_service.py # Cache statistics
│ │ │
│ │ ├── discovery/
│ │ │ └── discovery_service.py # Tool discovery service
│ │ │
│ │ ├── execution/
│ │ │ ├── secure_tool_execution_service.py # With guardrails
│ │ │ ├── tool_execution_service.py # Basic execution
│ │ │ └── execution_utils.py # Utilities
│ │ │
│ │ ├── oauth/ # 🔐 OAuth 2.0/2.1 service
│ │ │ ├── oauth_service.py # Core OAuth implementation
│ │ │ ├── token_manager.py # Token caching & refresh
│ │ │ ├── integration.py # Client integration
│ │ │ ├── models.py # OAuth data models
│ │ │ ├── validation.py # Scope & token validation
│ │ │ └── metrics.py # OAuth metrics tracking
│ │ │
│ │ ├── server/
│ │ │ ├── server_listing_service.py # List all servers
│ │ │ └── server_info_service.py # Server details
│ │ │
│ │ └── timeout/
│ │ └── timeout_manager.py # Timeout management
│ │
│ ├── bad_mcps/ # 🧪 Test MCP servers (security testing)
│ │ ├── echo_mcp.py # Simple echo server
│ │ ├── echo_oauth_mcp.py # OAuth header testing server
│ │ ├── bad_mcp.py # Multiple attack vectors
│ │ ├── bad_output_mcp.py # Malicious output testing
│ │ ├── command_injection_mcp.py # Command injection scenarios
│ │ ├── credential_theft_mcp.py # Credential theft attacks
│ │ ├── mpma_mcp.py # Multi-parameter manipulation
│ │ ├── path_traversal_mcp.py # Path traversal attacks
│ │ ├── prompt_injection_mcp.py # Prompt injection attacks
│ │ ├── rce_mcp.py # Remote code execution
│ │ ├── resource_exhaustion_mcp.py # DoS/resource exhaustion
│ │ ├── schema_poisoning_mcp.py # Schema manipulation
│ │ ├── session_management_mcp.py # Session attacks
│ │ ├── ssrf_mcp.py # Server-side request forgery
│ │ ├── tool_poisoning_mcp.py # Tool definition attacks
│ │ └── unauthenticated_access_mcp.py # Auth bypass scenarios
│ │
│ └── example_enkrypt_mcp_config.json # Example configuration
│
├── observability/ # Observability stack -- TWO backends, run one
│ ├── docker-compose.opensearch.yml # OpenSearch/Data Prepper stack (PRIMARY; OTLP host :4317/:4318)
│ ├── docker-compose.grafana.yml # Grafana/Prometheus/Loki/Jaeger stack (legacy; OTLP host :4327/:4328)
│ ├── .env.opensearch.example # env template -> copy to .env.opensearch
│ ├── .env.grafana.example # env template -> copy to .env.grafana
│ ├── README.opensearch.md # OpenSearch stack operator guide
│ ├── README.md # Grafana stack operator guide
│ ├── emit_dummy_telemetry.py # synthetic OTLP emitter (dashboard/monitor verification)
│ ├── opensearch/ # bootstrap.sh + ISM policy / templates / monitors / channel
│ │ ├── bootstrap.sh # idempotent installer (policy, templates, data streams, monitors, local user)
│ │ ├── policies/ # gateway_telemetry_policy.json (7d hot->delete, prio 1000)
│ │ ├── templates/ # gateway-{metrics,traces,logs}-elastic-template.json (SS4O data streams)
│ │ ├── monitors/ # 01..09 bucket-level alerting monitors
│ │ └── notification_channels/ # slack-mcpgw-alerts.json
│ ├── opensearch_dashboards/ # saved-objects.ndjson + gateway-dashboards.ndjson (13 viz + 3 dashboards)
│ ├── data_prepper/ # pipelines.yaml + render_pipelines.py (+ unit tests) + config
│ ├── otel_collector/ # otel-collector-config.yaml (Grafana) + .opensearch.yaml (OpenSearch)
│ ├── grafana/ # Grafana dashboards + alert provisioning
│ ├── prometheus/ # Prometheus scrape config
│ ├── loki/ # Loki log-aggregation config
│ └── promtail/ # Promtail config
│
│ NOTE: the apiaas repo is canonical for the OpenSearch resources;
│ observability/opensearch{,_dashboards}/ is a vendored mirror kept in
│ sync by scripts/install/opensearch/sync-check.sh (see apiaas repo).
│
├── docs/ # Documentation
│ ├── claude/
│ │ └── k8s-deployment-runbook.md # ⭐ EKS dev/prod deploy runbook (manual today)
│ ├── secure-mcp-gateway-manifest.yaml # LIVE dev manifest (gitignored -- holds creds)
│ └── secure-mcp-gateway-manifest-example.yaml # tracked template for the above
├── pyproject.toml # Python project config (PEP 621)
├── requirements.txt # Pinned runtime deps
├── Dockerfile / Dockerfile-Base # Container build
├── README.md # Main documentation
├── CHANGELOG.md # Version history
├── CLI-Commands-Reference.md # CLI documentation
└── API-Reference.md # API documentation
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed c8512a3cb748
- 1mo ago First seen · 1,580 lines · 13,326 tokens per session scan A 0b96c60d0a04
secure-mcp-gateway CLAUDE.md is an instructions file published in the GitHub repository enkryptai/secure-mcp-gateway (58 stars, last pushed 5d ago), licensed Apache-2.0. It adds 13,326 tokens to every session, about $0.0533 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
vscode model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.