Log inSubmit server
Infrawise logo

Infrawise

by Sidd27

Analyzes AWS infrastructure, databases, and code to expose drift, security, indexing, and performance findings through MCP.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Infrawise

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "sidd27-infrawise": {
      "command": "npx",
      "args": [
        "-y",
        "infrawise"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— we haven't run this command through our automated sandbox yet. Not a red flag; we're working through the catalog.

Overview

Sidd27/infrawise connects AI coding assistants to an analyzed view of an AWS-backed codebase, infrastructure, and database schemas. It exposes read-only MCP tools for infrastructure graphs, service metadata, table structures, function analysis, and database index recommendations. Analysis is cached, refreshed from local code and cloud metadata, and served through stdio or HTTP. Reach for it when an assistant needs concrete AWS relationships, schema details, event shapes, or implementation-ready index changes instead of inferring them from source code alone.

Use cases

  • Inspect AWS service relationships before changing application code
  • Find missing database indexes and generate migration statements
  • Analyze Lambda handlers for event shapes, scans, N+1 access, and IAM gaps
  • Review queues, topics, APIs, and triggers for configuration risks
  • Give coding assistants current schema and infrastructure context

Key features

  • AWS infrastructure graph and service overviews
  • Lambda, API, queue, topic, event, and storage inspection
  • DynamoDB, PostgreSQL, MongoDB, and MySQL index suggestions
  • Schema and code relationship analysis without row or secret values
  • Cached analysis with incremental code refreshes
  • Read-only MCP tools over stdio or HTTP

Tools22

What Infrawise lets your agent do. Expand a tool for its parameters.

Self-reported~691 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

17/18 checks healthy over the last 79d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 9d ago via OSV.dev · infrawise (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars22Stargazers on the source repository.
npm downloads1kPackage downloads in the last 30 days.
Last commit5d agoMost recent push to the default branch.
Tools exposed22Callable tools this server registers over MCP.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How Infrawise works

What Sidd27/infrawise does

Sidd27/infrawise gives MCP clients deterministic context about a project's code, AWS resources, and database schemas. It identifies infrastructure relationships and findings such as IaC drift, missing indexes, full-table scans, N+1 access patterns, security configuration gaps, and performance risks.

The server focuses on inspection and recommendations rather than provisioning or mutation. Its tools are annotated as read-only. Results can include service counts, high-severity findings, graph relationships, resource configuration, and data-health details describing snapshot age and source status.

Supported resource coverage includes Lambda, API Gateway, SQS, SNS, EventBridge, S3, CloudWatch Logs, Secrets Manager, SSM Parameter Store, Cognito, Kinesis, MSK, ElastiCache, and CloudFront. Database-oriented inspection covers DynamoDB, PostgreSQL, MongoDB, and MySQL.

How it works

Run the CLI from the project directory. On first startup, it probes the environment, creates an infrawise.yaml configuration file, scans the codebase and AWS sources, and writes an MCP configuration file for the selected editor. The resulting snapshot is cached for 24 hours.

The stdio process can be spawned by an editor without loading the snapshot immediately. The first infrastructure tool call loads the cache; an absent or expired cache triggers a fresh analysis, while concurrent initial calls share that work. Running infrawise start before opening the editor prepares the cache and avoids that initial wait.

Local code changes are tracked after the snapshot loads. TypeScript refreshes scan changed files and transitive importers, Python refreshes scan changed Python files, and file additions or deletions update the snapshot. IaC extraction, graph construction, and analyzers continue to evaluate the complete snapshot so cross-resource findings remain available. Cloud metadata remains cached.

Setup and configuration

Node.js 22 or newer is required. Install the package globally with:

Terminal
npm install -g infrawise

Or run it without a global installation:

Terminal
npx infrawise start --claude

For Claude Code, use infrawise start --claude; it creates or merges .mcp.json and opens Claude Code. Cursor uses infrawise start --cursor, which writes .cursor/mcp.json. VS Code uses infrawise start --vscode, which writes .vscode/mcp.json for Copilot agent mode. Running infrawise start without an editor flag writes a general .mcp.json and exits.

The server can also use HTTP transport:

bash
infrawise serve

This exposes the MCP endpoint at http://localhost:3000/mcp. For a manually configured client, stdio can be started with infrawise serve --stdio --config /path/to/infrawise.yaml.

Tools and capabilities

  • Inspect an overall infrastructure snapshot or the full resource graph.
  • Read table and collection schemas without retrieving row data.
  • Analyze a named function, including scans, N+1 patterns, trigger event shapes, IAM gaps, and unresolved Lambda links. An optional file selector distinguishes same-named files.
  • Generate exact DynamoDB GSI settings, PostgreSQL CREATE INDEX statements, MongoDB createIndex commands, and MySQL ALTER TABLE ADD INDEX statements.
  • Inspect queues, topics, APIs, secrets metadata, parameters, Lambda functions, event rules, buckets, logs, stack outputs, Cognito pools, streams, caches, and CloudFront distributions.
  • Preserve sensitive boundaries: secret values, parameter values, log message bodies, and database row data are not returned.

Limitations and notes

Infrawise is an infrastructure analysis layer, not an agent framework, provisioning system, observability platform, or cloud management dashboard. Its recommendations are based on the scanned code, IaC, schemas, and cloud metadata available to the analysis.

Resource relationships may include an explanation when a Lambda or other source cannot be resolved. Cached analysis can become stale; responses expose data-health information, and stack outputs parsed from local IaC can carry staleness flags for orphaned cdk.out templates. The server does not change infrastructure because its MCP tools are read-only.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Infrawise

Sidd27/infrawise is an MCP server that connects AI coding assistants to analyzed AWS infrastructure, application code, and database schemas. Its main tools inspect services, graphs, tables, Lambda functions, events, queues, and storage, while generating exact index recommendations for DynamoDB, PostgreSQL, MongoDB, and MySQL.

Install it with Node.js 22 or newer using `npm install -g infrawise`, or run `npx infrawise start --claude` without a global installation. Starting from the project directory generates configuration and scans the codebase, AWS services, and database sources.

Yes, Sidd27/infrawise works with Cursor. Run `infrawise start --cursor` from the project directory; it writes or merges `.cursor/mcp.json` and opens Cursor with the MCP tools available.

Run `infrawise serve` to start the HTTP MCP endpoint at `http://localhost:3000/mcp`, then add that URL to the client’s MCP configuration. Stdio is also available with `infrawise serve --stdio --config /path/to/infrawise.yaml`.

No, it does not return secret values, SSM parameter values, log message bodies, or database row data. Its inspection tools expose metadata such as names, schemas, configuration, inferred key names, and relationships.

Related MCP servers

All in Cloud Platforms All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewInfrawise AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/sidd27-infrawise?style=directory)](https://allmcps.com/mcp/sidd27-infrawise)
HTML Embed
<a href="https://allmcps.com/mcp/sidd27-infrawise"><img src="https://allmcps.com/api/badge/sidd27-infrawise?style=directory" alt="Infrawise on AllMCPs" /></a>