S2T Accelerators MCP Server

36 production-ready tools for AI agents -- covering AWS infrastructure, security compliance, AI workflows, stakeholder interviews, agent orchestration, and ACI governance.

Works with Claude Code, Claude Desktop, Cursor, Windsurf, OpenAI ChatGPT, and any MCP-compatible client.

Quick Start

Option 1: npx (no install)

S2T_API_KEY=sk_live_your_key npx s2t-mcp-accelerators

Option 2: Global install

npm install -g s2t-mcp-accelerators

Get an API Key

Visit https://dev.s2tconsulting.com/ai-sales/purchase.html to create an account and get your API key. A free tier is available.

Configuration

Claude Code / Claude Desktop

Add to your MCP config file:

Claude Code: ~/.mcp.json (or project-level .mcp.json) Claude Desktop (macOS/Linux): ~/.config/Claude/claude_desktop_config.json Claude Desktop (Windows): %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "s2t-accelerators": {
      "command": "npx",
      "args": ["-y", "s2t-mcp-accelerators"],
      "env": {
        "S2T_API_KEY": "sk_live_your_api_key_here"
      }
    }
  }
}

Cursor / Windsurf

Add to your MCP settings with the same structure as above.

Streamable HTTP (Remote)

For web-based clients that support Streamable HTTP transport:

URL: https://mcp.s2tconsulting.com/mcp
Header: X-S2T-API-Key: sk_live_your_api_key_here

Environment Variables

| Variable | Description | Default | |----------|-------------|---------| | S2T_API_KEY | Your S2T API key (required) | -- | | S2T_API_URL | Custom API endpoint | Production API | | PORT | HTTP server port (HTTP mode only) | 3001 |

All 36 Tools

Infrastructure & Cloud (4 tools)

| Tool | Description | |------|-------------| | s2t_generate_cloudformation | Generate production-ready SAM/CloudFormation templates from natural language | | s2t_generate_dynamodb_design | Generate optimal DynamoDB single-table designs from entity and access patterns | | s2t_validate_oauth | Validate OAuth 2.0 configs for Google, Microsoft, GitHub, QuickBooks | | s2t_check_data_lake_readiness | Assess data lake architecture readiness with scoring and recommendations |

Security & Compliance (3 tools)

| Tool | Description | |------|-------------| | s2t_validate_iam_policy | Validate IAM policies for security best practices with scoring (0-100) | | s2t_validate_mfa_compliance | Check IAM users/roles for MFA compliance with remediation steps | | s2t_validate_cli_readiness | Validate CLI tool availability and API key health |

AI & Embeddings (2 tools)

| Tool | Description | |------|-------------| | s2t_embed | Generate vector embeddings via Amazon Titan for RAG and semantic search | | s2t_analyze_error_patterns | Analyze error logs to identify patterns, root causes, and trends |

Agent Orchestration (7 tools)

| Tool | Description | |------|-------------| | s2t_classify_action_risk | Classify action risk level (LOW/MEDIUM/HIGH/CRITICAL) for agent safety | | s2t_route_task_to_agent | Route tasks to optimal agents via semantic matching across 54 domains | | s2t_predict_system_issues | Predict system issues from budget trends, error rates, dependencies | | s2t_attempt_auto_recovery | Match errors against known patterns and suggest recovery steps | | s2t_execute_with_resilience | Configure retry + circuit breaker patterns for operations | | s2t_manage_agent_memory | Persistent agent memory with namespace isolation across sessions | | s2t_submit_agent_task | Submit tasks to SQS FIFO queue for async multi-agent execution |

Distributed Systems (2 tools)

| Tool | Description | |------|-------------| | s2t_create_trace_context | Generate W3C Trace Context IDs for distributed tracing | | s2t_acquire_file_lock | File-based mutex locks with stale lock detection |

Interview & Engagement (4 tools)

| Tool | Description | |------|-------------| | s2t_interview_create | Create and start a structured stakeholder interview session | | s2t_interview_message | Send a response and receive the next adaptive interview question | | s2t_interview_summary | Generate structured findings from a completed interview session | | s2t_interview_list | List active and completed interview sessions |

Platform (2 tools)

| Tool | Description | |------|-------------| | s2t_catalog | List all available accelerators with capabilities and pricing | | s2t_usage | Get your current API usage statistics and remaining quota |

ACI Governance (12 tools)

Tools for Autonomous Corporate Intelligence -- enterprise AI agent governance with immutable audit trails.

| Tool | Description | |------|-------------| | aci_classify_decision | Classify agent actions as APPROVE/ESCALATE/BLOCK with audit logging | | aci_financial_gate | Estimate financial impact and check budget gates before execution | | aci_compliance_check | Evaluate actions against SOC2, GDPR, HIPAA, PCI-DSS frameworks | | aci_route_domain | Route governance tasks to domain expert agents with sensitivity tagging | | aci_parallel_review | Initiate parallel governance review across multiple domain agents | | aci_synthesize_reviews | Collect and synthesize results from parallel review sessions | | aci_log_decision | Record governance decisions in the immutable audit log | | aci_recall_precedent | Semantic search over past governance decisions for precedent | | aci_record_outcome | Record actual outcomes to calibrate future classification accuracy | | aci_estimate_blast_radius | Estimate blast radius: affected systems, users, recovery time | | aci_generate_rollback | Generate step-by-step rollback plans for high-risk operations | | aci_governance_health | Aggregate governance health metrics and calibration status |

Pricing

| Tier | Price | Rate Limit | Monthly Limit | |------|-------|------------|---------------| | Free | $0 | 10/min | 100 calls | | Developer | $29/mo | 60/min | 5,000 calls | | Business | $99/mo | 300/min | 50,000 calls | | Enterprise | $299/mo | 1,000/min | Unlimited |

All tiers include access to all 36 tools. ACI governance tools are included in Business tier and above.

Transports

| Transport | Entry Point | Protocol | |-----------|-------------|----------| | stdio | s2t-mcp-accelerators | MCP over stdin/stdout | | HTTP | s2t-mcp-accelerators-http | Streamable HTTP + legacy SSE |

The stdio transport is used by default with npx. The HTTP transport can be started with:

S2T_API_KEY=sk_live_your_key npx s2t-mcp-accelerators-http
# Listening on http://localhost:3001/mcp

Local Development

git clone https://github.com/S2TConsulting/accelerators-mcp.git
cd accelerators-mcp
npm install
npm run build
npm test

MCP Inspector

Test the server interactively:

npx @modelcontextprotocol/inspector node dist/index.js

Running Locally

export S2T_API_KEY="sk_live_your_api_key"

# Stdio mode
npm start

# HTTP mode
npm run start:http

Testing

npm test              # Run tests once
npm run test:watch    # Watch mode
npm run test:coverage # Coverage report

Contributing

1. Fork the repo and create a feature branch
2. Install dependencies: npm install
3. Make changes in src/
4. Add tests for new functionality
5. Run the test suite: npm test
6. Submit a PR with a clear description

Support

• Website: https://www.s2tconsulting.com
• Documentation: https://www.s2tconsulting.com/accelerators
• Email: [email protected]
• Issues: https://github.com/S2TConsulting/accelerators-mcp/issues

License

MIT -- See LICENSE file for details.