Open Source MCP Servers for AWS
Infrastructure · tested 2026-08-14 · by the Hlido desk, not the vendor
In short: AWS's own suite of MCP servers, documented like AWS documents things — comprehensive, categorised and institutionally maintained; the catch is that 'it' is a collection, so a single score flattens servers that differ widely in maturity.
Quick answer
Open Source MCP Servers for AWS scores 83/100 (STEADY) on Hlido’s independent, hands-on test (reviewed 2026-08-14). STEADY and toward the upper end because provenance, documentation breadth, categorised organisation, named client support and working samples are all present and verifiable, and vendor continuity is effectively assured. Pricing: Paid.
This is the official AWS Labs MCP effort: a suite of specialised servers organised into named categories — migration and modernization, infrastructure and deployment, AI/ML, data and analytics, developer tools, integration and messaging, cost and operations, healthcare and lifesciences — plus getting-started material, installation guidance, samples (Bedrock Knowledge Bases, Nova Canvas, Step Functions) and a 'vibe coding tips' section. The documentation is public, versioned in the open, multilingual at the shell level, and backed by a vendor whose continuity is not in question. Named client support spans Kiro, Cline, Cursor, Windsurf and Claude Desktop. For a team already on AWS, this is the default starting point and the provenance alone settles most of the trust question. The honest limitation is structural, and it applies to the review as much as to the product. What is being scored is a catalogue, not a single agent. The individual servers within it vary considerably in maturity, tool surface and update cadence, and the landing documentation is deliberately introductory — it explains what MCP is before it explains what any particular server does. A buyer evaluating, say, the cost-and-operations server needs to descend into that server's own documentation, and the aggregate score here says little about it. Read this as a verdict on the suite's provenance, organisation and documentation quality, which are strong, rather than on any one server's depth.
Why STEADY
STEADY and toward the upper end because provenance, documentation breadth, categorised organisation, named client support and working samples are all present and verifiable, and vendor continuity is effectively assured. Not in the top band because the reviewed surface is an introductory catalogue rather than a product: it does not publish a consolidated tool inventory, per-server maturity signals or version information, so the depth of any individual server remains unverified from here.
What we saw
4 screenshots captured by the Hlido engine during the reviewed run (run-d32a0ab121b409d8-awslabs-github-io). Our own captures — not vendor marketing material.
What it does well
- First-party AWS provenance — continuity, security process and maintenance are not open questions
- Servers organised into clear functional categories rather than dumped as an undifferentiated list
- Public documentation with installation guidance and no login wall
- Working samples published (Bedrock Knowledge Bases, Nova Canvas, Step Functions)
- Broad named client support: Kiro, Cline, Cursor, Windsurf, Claude Desktop
- Open source, developed in the open under awslabs
What it fails at
- It is a suite, not a product — a single evaluation cannot represent servers that differ widely in maturity
- No consolidated tool inventory or per-server version and maturity signals on the landing surface
- Landing documentation is introductory (explaining MCP itself) rather than reference material
- Strongly AWS-centric by design; little value outside that ecosystem
- No stability or support-tier statement distinguishing production-ready servers from experimental ones
Red flags
- No public signal distinguishes production-ready servers from experimental ones within the suite, so 'AWS Labs' provenance can read as more uniform assurance than it actually provides.
- The aggregate nature of the surface means a favourable overall assessment should not be transferred to any individual server without checking that server directly.
Best for
- Teams already operating on AWS who want first-party MCP access to AWS services
- Engineers who weight provenance and continuity heavily in tooling decisions
- Organisations needing documentation and an open-source supply chain they can audit
- Anyone starting with MCP on AWS — this is the correct default entry point
Not recommended for
- Multi-cloud or non-AWS environments, where the suite has little to offer
- Buyers needing per-server maturity or support-tier guarantees before production use
- Teams wanting a single focused server rather than a catalogue to navigate
Pricing & access
- ModelPaid
Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-08-13.
Compared to
-
MikroMCP
focused-safety-vs-vendor-breadth
MikroMCP is one focused server for physical network gear with an explicit safety model; the AWS suite is many servers for cloud services with vendor backing. Different infrastructure, and the AWS suite does not publish an equivalent per-server change-safety story.
-
nable
opinionated-workflow-vs-first-party-primitives
nable covers cloud cost across 14 providers including AWS, with an opinionated read-only-plus-PR workflow; the AWS suite's cost-and-operations servers are first-party but leave the workflow to you.
Agent relevance
API CLI MCP SDK Behavioral-testable
Agentic-Commerce Readiness 67/100 · INTEGRABLE
Independent readiness for agent delegation & transaction. How it’s scored · check live
A catalogue of MCP servers with installation guidance and broad named client support. Individual servers expose their own tool surfaces, so agent-addressability is high in principle but must be established per server — the landing surface publishes no consolidated inventory to reason from.
Agent-friendly score: 8/10
Score over time
The longitudinal record — every point is the score as published on that date. Raw series.
Evidence
- Homepage publicly accessible and value proposition clearly stated — source (2026-08-14) verified
- Pricing page discoverable in 2 clicks from homepage — source (2026-08-14)
- Documentation or live demo accessible without login — source (2026-08-14) verified
- Integration list or supported frameworks documented — source (2026-08-14) verified
- Authentication / data handling claims publicly stated — source (2026-08-14)
scorecard.json · transparency passport · registry · methodology
Embed this trust badge
Markdown
[](https://hlido.eu/check/?agent=awslabs-mcp)HTML
<a href="https://hlido.eu/check/?agent=awslabs-mcp"><img src="https://hlido.eu/badge/awslabs-mcp.svg" alt="Hlido trust score"></a>