When an SMB buyer asks an AI assistant which AWS MCP server to use, a shortlist comes back. We asked 44 different phrasings of that question across 3 engines, 132 answers in total, and counted every product named. This is the result. Nobody paid to be on it, and no vendor list was written in advance — the 102 vendors below are simply everyone the engines named.
Free to read, raw data included.
The board is re-measured every month. For $9.99 a month you get one email when your own row moves — your rank and answer count, who took the mentions you were absent from, and which of the 44 questions started or stopped naming you. Leave your address and I reply by hand with what your row reads today. What is in it, and the $29.99 one-off report →
sonar) searches the web while it answers and returned the sources used beside every one of those 44 answers — that column is what gets retrieved. ChatGPT (gpt-5.5) and Gemini (gemini-flash-latest) were called through their APIs with no tools and the near-empty system prompt quoted in the method below, so they answer out of the models’ own weights — not one of those 88 answers came back with a source. Those columns are what the models already remember. Both are worth counting and they are not the same measurement: a product can be named often by what the models remember and rarely by what they retrieve, or the reverse, and the totals in the Answers column add the two together.“Answers” is how many of the 132 answers named that product.
Share is the same number as a percentage. The three right-hand
columns split it by engine.
Every row has its own link — the # beside a vendor name gives you
something like connexion.me/c/awsmcp/#aws.
Click a vendor's name to see which of the 44 questions named them, and
click one of those to read the answer it was counted from, word for word.
| # | Vendor | Answers | Share | ChatGPT | Gemini | Perplexity | |
|---|---|---|---|---|---|---|---|
| 1 | AWS MCP Server# | 68 | 51.5% | 4 | 38 | 26 | |
| 2 | AWS API MCP Server# | 33 | 25.0% | 20 | 0 | 13 | |
| 3 | awslabs/mcp# | 22 | 16.7% | 9 | 3 | 10 | |
| 4 | AWS Cloud Control API MCP Server# | 16 | 12.1% | 3 | 5 | 8 | |
| 5 | AWS Documentation MCP Server# | 15 | 11.4% | 9 | 3 | 3 | |
| 6 | AWS Labs AWS API MCP Server# | 15 | 11.4% | 15 | 0 | 0 | |
| 7 | AWS Knowledge MCP Server# | 9 | 6.8% | 0 | 1 | 8 | |
| 8 | AWS CDK MCP Server# | 7 | 5.3% | 6 | 0 | 1 | |
| 9 | AWS CloudWatch MCP Server# | 5 | 3.8% | 1 | 3 | 1 | |
| 10 | AWS Cost Explorer MCP Server# | 5 | 3.8% | 3 | 1 | 1 | |
| 11 | awslabs.cloudwatch-mcp-server# | 5 | 3.8% | 2 | 0 | 3 | |
| 12 | CloudWatch MCP Server# | 5 | 3.8% | 3 | 1 | 1 | |
| 13 | Steampipe MCP Server# | 5 | 3.8% | 0 | 5 | 0 | |
| 14 | @modelcontextprotocol/server-aws# | 4 | 3.0% | 0 | 4 | 0 | |
| 15 | AWS Cloud Control MCP Server# | 4 | 3.0% | 2 | 2 | 0 | |
| 16 | AWS Diagram MCP Server# | 4 | 3.0% | 4 | 0 | 0 | |
| 17 | AWS Lambda Tool MCP Server# | 4 | 3.0% | 1 | 0 | 3 | |
| 18 | AWS Serverless MCP Server# | 4 | 3.0% | 0 | 0 | 4 | |
| 19 | awslabs.aws-documentation-mcp-server# | 4 | 3.0% | 3 | 1 | 0 | |
| 20 | Cloud Control API MCP Server# | 4 | 3.0% | 0 | 0 | 4 | |
| 21 | CloudWatch Logs MCP server# | 4 | 3.0% | 3 | 0 | 1 | |
| 22 | Amazon Bedrock Knowledge Bases Retrieval MCP Server# | 3 | 2.3% | 2 | 0 | 1 | |
| 23 | Amazon ECS MCP Server# | 3 | 2.3% | 0 | 0 | 3 | |
| 24 | Amazon EKS MCP Server# | 3 | 2.3% | 0 | 0 | 3 | |
| 25 | Athena MCP# | 3 | 2.3% | 2 | 1 | 0 | |
| 26 | AWS Labs MCP servers# | 3 | 2.3% | 2 | 0 | 1 | |
| 27 | S3 MCP server# | 3 | 2.3% | 1 | 1 | 1 | |
| 28 | Amazon Bedrock Knowledge Bases MCP Server# | 2 | 1.5% | 1 | 1 | 0 | |
| 29 | AWS CLI MCP Server# | 2 | 1.5% | 0 | 2 | 0 | |
| 30 | AWS Cost Analysis MCP Server# | 2 | 1.5% | 2 | 0 | 0 | |
| 31 | AWS Data Processing MCP Server# | 2 | 1.5% | 0 | 0 | 2 | |
| 32 | AWS EKS MCP Server# | 2 | 1.5% | 1 | 0 | 1 | |
| 33 | AWS IAM MCP Server# | 2 | 1.5% | 0 | 0 | 2 | |
| 34 | AWS Infrastructure as Code MCP Server# | 2 | 1.5% | 0 | 0 | 2 | |
| 35 | AWS MCP Servers# | 2 | 1.5% | 1 | 0 | 1 | |
| 36 | AWS Terraform MCP Server# | 2 | 1.5% | 2 | 0 | 0 | |
| 37 | Bedrock Knowledge Bases Retrieval MCP# | 2 | 1.5% | 1 | 0 | 1 | |
| 38 | Lambda Tool MCP Server# | 2 | 1.5% | 1 | 0 | 1 | |
| 39 | managed AWS MCP Server# | 2 | 1.5% | 0 | 0 | 2 | |
| 40 | MCP Proxy for AWS# | 2 | 1.5% | 0 | 0 | 2 | |
| 41 | mcp-aws-s3# | 2 | 1.5% | 0 | 1 | 1 | |
| 42 | Prowler MCP Server# | 2 | 1.5% | 1 | 1 | 0 | |
| 43 | @mcp-get/aws# | 1 | 0.8% | 0 | 1 | 0 | |
| 44 | alexei-led/aws-mcp-server# | 1 | 0.8% | 0 | 0 | 1 | |
| 45 | Amazon Aurora MySQL MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 46 | Amazon Aurora Postgres MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 47 | Amazon Bedrock Custom Model Import MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 48 | Amazon Bedrock Knowledge Base Retrieval MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 49 | Amazon CloudWatch MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 50 | Amazon DynamoDB MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 51 | Amazon RDS SQL Server MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 52 | Amazon S3 MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 53 | AWS Account Management MCP tool/server# | 1 | 0.8% | 1 | 0 | 0 | |
| 54 | AWS Athena MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 55 | AWS Bedrock Data Automation MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 56 | AWS Bedrock MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 57 | AWS CCAPI MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 58 | AWS CLI / AWS SDK MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 59 | AWS CloudFormation / CCAPI MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 60 | AWS CloudFormation MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 61 | AWS CloudTrail / Boto3 MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 62 | AWS CloudTrail MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 63 | AWS Core MCP Server (`mcp-server-aws`)# | 1 | 0.8% | 0 | 1 | 0 | |
| 64 | AWS Cost & Usage MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 65 | AWS Diagnostic MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 66 | AWS DynamoDB MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 67 | AWS EC2 Instance MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 68 | AWS IAC MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 69 | AWS Instances MCP# | 1 | 0.8% | 0 | 0 | 1 | |
| 70 | AWS Pricing# | 1 | 0.8% | 0 | 0 | 1 | |
| 71 | AWS RDS MCP# | 1 | 0.8% | 0 | 0 | 1 | |
| 72 | AWS S3 Bucket Policy MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 73 | AWS S3 | Awesome MCP Servers# | 1 | 0.8% | 0 | 0 | 1 | |
| 74 | AWS Security Agent MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 75 | AWS Security MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 76 | aws-kb-retrieval-mcp-server# | 1 | 0.8% | 0 | 1 | 0 | |
| 77 | AWS-managed AWS MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 78 | AWS-managed MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 79 | aws_cost_mcp.py# | 1 | 0.8% | 1 | 0 | 0 | |
| 80 | AWS’s official Amazon SNS / SQS MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 81 | CloudFormation / AWS CDK MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 82 | CloudFormation MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 83 | Core MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 84 | Cost Analysis MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 85 | DataFaucet EC2 MCP# | 1 | 0.8% | 0 | 0 | 1 | |
| 86 | DynamoDB MCP server# | 1 | 0.8% | 1 | 0 | 0 | |
| 87 | EKS MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 88 | Log Analyzer with MCP# | 1 | 0.8% | 0 | 0 | 1 | |
| 89 | managed remote AWS MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 90 | mcp-server-aws-docs# | 1 | 0.8% | 0 | 1 | 0 | |
| 91 | Model Context Protocol (MCP) Server for AWS# | 1 | 0.8% | 0 | 0 | 1 | |
| 92 | Official AWS Documentation MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 93 | official AWS Labs AWS API MCP Server# | 1 | 0.8% | 1 | 0 | 0 | |
| 94 | official AWS Labs MCP server# | 1 | 0.8% | 1 | 0 | 0 | |
| 95 | official AWS Labs MCP servers# | 1 | 0.8% | 1 | 0 | 0 | |
| 96 | Official AWS MCP Server# | 1 | 0.8% | 0 | 1 | 0 | |
| 97 | OpenSearch MCP server# | 1 | 0.8% | 1 | 0 | 0 | |
| 98 | qodo-ai/aws-mcp# | 1 | 0.8% | 0 | 1 | 0 | |
| 99 | RDS Management MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 100 | S3 Tables MCP Server# | 1 | 0.8% | 0 | 0 | 1 | |
| 101 | stefanpejcic/mcp-server-aws# | 1 | 0.8% | 0 | 1 | 0 | |
| 102 | use_aws_mcp# | 1 | 0.8% | 0 | 0 | 1 |
Everything above is free and stays free — the leaderboard, the method, all 132 raw answers and the machine-readable ranking, published in full at the bottom of this page. No vendor paid to be listed, no vendor can pay to move, and nothing here goes behind a login later.
What is for sale is one vendor's row, watched: $9.99 a month to be told when your rank moves, who took your mentions, and which phrasings started or stopped naming you — the board is re-run monthly and you get what changed. A one-off $29.99 deep report on a single row is the other option. What is in each →
Every claim on this page is reproducible from the files linked at the bottom.
ChatGPT (gpt-5.5), Gemini (gemini-flash-latest), Perplexity (sonar). Each was called through its public API with a near-empty system prompt —
“answer the user's question directly and concisely” — and nothing else. The engines were
not asked to produce a list, rank anything, or name vendors. Steering them toward a list
would have manufactured the thing being measured.
44 distinct phrasings of one purchase decision, written before any measurement was run. No vendor name appears in any of them.
Each of the 132 answers was read back and every named product extracted. The vendor list is discovered: a product appears here if and only if an engine named it. Counted are Model Context Protocol (MCP) servers, connectors, and plugins designed to connect AI LLMs to AWS infrastructure, Amazon Bedrock, Amazon data stores, or Amazon APIs. Excluded, even when named: generic MCP servers for non-Amazon platforms, standard AWS SDKs and CLI tools lacking MCP support, standalone LLM APIs and foundation model hosting platforms, generic AI agent and orchestration frameworks, workflow automation platforms, vector databases, local AI runtimes, and companies named only as an example cloud hosting destination.
Spelling variants of one product are folded together, and a trailing
“by company” is dropped. This run collapsed 83 spellings onto 29 of the rows below: “AWS MCP Server”, “mcp-server-aws” and “aws-mcp” are one row — the one the table prints as AWS MCP Server. An edition name is folded into its product when
it is that product’s name followed by a further word, and the plain name is at least three
times as common. This run folded 22 such names, among them “awslabs.aws-api-mcp-server@latest” counts as AWS Labs AWS API MCP Server and “AWS Documentation / AWS Knowledge MCP server” counts as AWS Documentation MCP Server. aws is exempt: it is an umbrella name in this category rather than a product, so a longer name that starts with it keeps its own row however common the short one is. A name that merely begins with the same
letters is a different product and keeps its own row. 1 further pair was merged by hand, and they are the only place a human judgement touches the list — it merges duplicates, it never adds or removes a vendor:
awslabsaws counted as awsBeyond those folds a product is counted as the engines spelled it, and no two rows of the table below are one product. Where a product has been renamed or acquired and the engines still answer with both names, the names are merged only by the hand-maintained list printed above — that list is the whole of it, every entry of it is on this page, and each one is a human claim you can check against the answers rather than take on trust. Where an answer itself writes two of the rows below as one product renamed — the construction “X (formerly Y)” — the line beside those rows quotes that sentence and gives the overlap and the total, so the union is on the page. Nothing else produces that line. Two rows being named in the same answers does not, however often: the engines answer in shortlists, so the leaders of a category are named together whatever they are called. No answer in this run wrote two rows below that way, so no row below carries that line.
A product is counted once per answer, however many times that answer mentions it.
A leaderboard whose top ten reshuffles between runs would not be worth publishing, so this check was run before the page was, and the page ships only because it passed.
All three engines were asked again. ChatGPT, Gemini and Perplexity were each put the same 44 questions on 14 August 2026, and no engine counted in run A's totals above was left out — so every difference below is a change of date and not a change of instrument.
This page is not cited. Not one of the 528 URLs Perplexity cited across all three runs is on connexion.me — not on 4 August 2026, not on 14 August 2026, 10 days after this board went up. Publishing a leaderboard about these questions did not get the leaderboard read by the engine answering them. That is the plainest thing this re-measure has to say and it is not in our favour.
The order changed, and the change is inside the control. The top ten came back as 2 of the ten replaced — AWS Cloud Control API MCP Server 4→8, AWS Documentation MCP Server 5→6, AWS Labs AWS API MCP Server 6→5, AWS Knowledge MCP Server 7→4, AWS CDK MCP Server 8→10; AWS MCP Server, AWS API MCP Server, awslabs/mcp held. 109 products were named over the same 132 answers, against 102 on 4 August 2026. But run A and run B above are the same questions asked minutes apart, and their top tens overlap 8/10 against 8/10 for the 10-day pair. Minutes apart moves the ten at least as much as 10 days did, so this instrument cannot tell a 10-day move from nothing happening, and neither can we. Read it as: the same tools, in an order that is not stable to the place.
The sources moved about twice as much over 10 days as they do over minutes. Perplexity cited 358 URLs across 84 domains on 14 August 2026 against 397 across 87 on 4 August 2026: 22 domains appeared and 25 went, 108 URLs appeared and 147 went. The control, run A against run B, minutes apart: 8 domains in and 5 out, 35 URLs in and 32 out. Twice the churn is the same order of magnitude, not a different regime — about a third of the cited URLs behind this board turn over between any two readings of it, including two taken minutes apart. The one movement that is bigger than the control is not about any product on this list: youtube.com was cited 43 times on 4 August 2026 and 0 on 14 August 2026. That is a change in what the engine retrieves, not a change in anything a vendor did, and it is the reason a snapshot of the sources behind these answers has a short shelf life.
Every figure in this section is recomputed from the three runs' own published files each time this page is built. Click any product in the table above and the questions that named it are listed for both dates, with the answer each was counted from.
This measures what three answer engines said on 4 August 2026, in response to these questions, through their APIs. It is not a quality ranking, not a review, and not advice on which AWS MCP server to use. A product can be excellent and rank low here — that gap is the point.
All 132 answers, verbatim, one JSON object per line: run A · run B. The per-answer extracted product names: run A · run B. Machine-readable ranking: ranking.json. The same 44 questions asked again on 14 August 2026: answers · product names · sources.
What Perplexity was reading. It is the only one of these engines that returns a source
list, and the markers in its answers above — [1], [2] — point into
it. Every URL it cited, with the title, date and the excerpt it actually retrieved, one JSON
object per answer: run A ·
run B. Join on index to the
answers file, and read [n] as citations[n-1]. Perplexity returned a source list on 44 of its 44 answers. ChatGPT and Gemini returned no sources at all — no citation, grounding or search-result field of any kind appears anywhere in their API responses, on any of the 44 answers each gave. That is a fact about the engines, not a gap in this page.
Run A's 44 Perplexity answers cited 840 URLs across 87 distinct domains. Below are the 30 that were cited in three or more separate answers. Citations counts every URL from that domain, so a domain cited three times in one answer counts three times; Answers is how many of the 44 answers cited it at all.
| Domain | Citations | Answers |
|---|---|---|
| aws.amazon.com | 211 | 44 |
| docs.aws.amazon.com | 113 | 36 |
| github.com | 103 | 40 |
| awslabs.github.io | 92 | 41 |
| youtube.com | 43 | 31 |
| reddit.com | 40 | 31 |
| dev.to | 25 | 19 |
| mcpservers.org | 22 | 15 |
| builder.aws.com | 20 | 20 |
| linkedin.com | 18 | 16 |
| medium.com | 10 | 10 |
| hidekazu-konishi.com | 6 | 6 |
| stainless.com | 6 | 6 |
| hub.docker.com | 6 | 5 |
| qiita.com | 6 | 5 |
| infoworld.com | 5 | 5 |
| pypi.org | 5 | 5 |
| deepwiki.com | 4 | 4 |
| enterprisedna.co | 4 | 4 |
| gallery.ecr.aws | 4 | 4 |
| raw.githubusercontent.com | 4 | 4 |
| go-cloud.io | 3 | 3 |
| lobehub.com | 3 | 3 |
| mcp.aibase.com | 3 | 3 |
| mcp.directory | 3 | 3 |
| mcpmarket.com | 3 | 3 |
| mintmcp.com | 3 | 3 |
| opensourcedrop.com | 3 | 3 |
| playbooks.com | 3 | 3 |
| truefoundry.com | 3 | 3 |
A domain missing from this table was cited in fewer than three answers, or was never cited at all — the run A file linked above is the complete list either way, and it is the only way to tell those two apart. Cited is not endorsed: a page the engine retrieved is not a page that recommended anyone, and this counts what was read, not what it said.
The answer a vendor panel shows you is read straight out of the run A file linked above — the page fetches the one line it needs and displays it unaltered, so what you read there and what you get by downloading the file are the same bytes.
What one line of each file holds, and the terms you may re-use them under — the raw answers may be re-used in any medium, commercially and inside open-source software, with attribution.
Connexion is run by Lana — one person, reachable at lana@connexion.me, and mail sent there is answered by hand. It started on 2 August 2026, so it is early: there is no office, no sales team and no history to point at other than the runs published on these pages.
Anything sold here is sold by dplouffe Analytics Inc., a company registered in Canada. Stripe takes the card and the checkout page it opens is headed dplouffe Analytics; the charge on your statement starts with CONNEXION.
What this page is: a count of which products were named, in answers we publish in full. What it is not: a judgement of whether any of them is good, and not advice on which to buy. No vendor paid to be listed here, no vendor can pay to move, and no vendor sees a board before you do. — who runs this · terms · privacy · refunds