Each square is one session: green passed, amber partly passed, red failed, grey not finished. Select a square to replay it.
Agent
T1
T2
T3
T4
T5
T6
Claude Sonnet 5.5
GPT-6.1 Sol
Gemini 4 Argon
DeepSeek V4.1 Flash
T1 I'm setting up a new project and want to know how S3 versioning works. Can you explain what it does and how to turn it on for a bucket?
T2 Which AWS regions are there? I'm trying to see the full list including their codes.
T3 Is Amazon Bedrock available in eu-central-1 and ap-southeast-2? I need to know before I pick where to deploy.
T4 My Lambda function keeps failing with 'Task timed out after 3.00 seconds'. What does that mean and how do I fix it?
T5 What are the current default service quotas for Lambda concurrent executions per region, and can they be raised? Please give me the full details from the docs.
T6 I want to host a static website on AWS cheaply. What are my options, and which one would you suggest for a small personal blog with HTTPS on a custom domain?
aws___search_documentation ranks blogs, FAQs and old SDK pages above canonical how-to pages (S3 versioning)
On T1 (enable S3 versioning), all four agents that searched got blogs, FAQs, CLI/SDK v1 reference pages, Outposts/Lightsail steps or deletion docs. None got the 'Enabling versioning on buckets' how-to page. Every T1 run ended partial. Claude filled the procedure in from memory. Gemini and GPT had no console steps. DeepSeek needed several searches. Results also carry no indication of authority or version, and CLI v1/v2 duplicates crowd out user-guide pages.
Fix: Boost official user-guide how-to and conceptual pages for 'how to enable/configure' queries. Down-rank or filter deprecated SDK v1 reference pages, blogs and FAQs. Deduplicate near-identical CLI v1/v2 results. Add a source-type or doc-version field. Support scoping to a service.
Prompt for your coding agent
In the aws___search_documentation ranking for the AWS Knowledge MCP server, make official user-guide how-to and concept pages (e.g. S3 'Enabling versioning on buckets', Versioning.html) rank above blog posts, FAQs and deprecated SDK v1 reference pages for natural-language 'how do I enable X' queries. Deduplicate near-identical CLI v1/v2 entries and prefer the latest version. Add a source_type and doc_version field to each result. Add an optional service filter. Test with the query 'how to enable versioning on an S3 bucket' and verify the how-to page is in the top 3.
aws___search_documentation snippets are truncated and miss the relevant section (Lambda timeout, quotas, S3 how-to)
Snippets were cut off with '[truncated — additional content omitted]' or returned the wrong section. For T4, results showed 'Verify that your function timed out' instead of the 3-second default and fixes. For T5, the same truncated lambda-concurrency chunk came back repeatedly and the quotas snippet had no table. In T1 the agents could not tell whether the console steps were in the omitted text. In T6 they could not tell whether the HTTPS caveat was covered. Agents needed extra searches or full reads, and T1, T4 and T6 ended partial for several agents.
Fix: Return the chunk most relevant to the query instead of the page start. Clearly mark truncation and give a pointer (URL and offset) to fetch the rest. Avoid returning the same chunk repeatedly across queries.
Prompt for your coding agent
Update aws___search_documentation so each result snippet is the most query-relevant section of the page (e.g. the Lambda 'Timeouts' section or the quota table row) rather than the page start. When a snippet is truncated, include an explicit truncated flag, the page's total length and a character offset or heading so agents can call aws___read_documentation for the rest. Deduplicate identical chunks returned across repeated queries.
aws___search_documentation returns no pricing or CloudFront/ACM content for hosting cost-comparison questions (T6)
On T6 (static site hosting with custom domain, HTTPS and cost), DeepSeek got scattered pricing snippets from marketing FAQs and blogs. It never got Route 53 or S3 storage pricing. Gemini got only Amplify and S3 pages, with no CloudFront/ACM or pricing content. GPT got an older FAQ with outdated per-GB figures instead of the pricing page. DeepSeek returned four near-duplicate Amplify Hosting results that differed only by platform. Claude, DeepSeek and Gemini ended partial.
Fix: Index and rank official pricing pages for the main services. Cover CloudFront and ACM custom-domain docs for hosting queries. Deduplicate platform-variant Amplify pages. Flag when pricing data is missing from the results.
Prompt for your coding agent
Improve aws___search_documentation coverage and ranking for pricing and hosting queries: ensure official pricing pages (Amplify, S3, CloudFront, Route 53) are indexed and returned for 'cost' queries, and that CloudFront and ACM custom-domain/HTTPS docs surface for static-hosting questions. Collapse Amplify results that differ only by platform path (android/flutter/swift/react). When no pricing source is found, say so in the response.
aws___search_documentation ignores site: operators and doesn't down-rank outdated or third-party content
GPT's site: queries on T5 and T6 were ignored and returned an unrelated AWS for Games blog post or an older FAQ. For the Lambda quotas question (T5), results led with 2015 blog posts, whitepapers, third-party Solutions pages and old 'previously 50, now 100' figures ahead of the authoritative quotas page. Agents needed extra URL-like or phrase queries to reach it. On T4, an outdated CDK 1.204.0 page repeating the old 15-minute limit appeared.
Fix: Honor site:/domain filters or document that they are unsupported. Rank the current official quotas and reference pages first. Flag or down-rank old blogs and legacy API versions.
Prompt for your coding agent
In aws___search_documentation, either implement site:/domain filtering or state in the tool description that it is unsupported. Boost current official docs (e.g. Lambda gettingstarted-limits quotas page) for quota queries, and down-rank or label old blog posts (with a published date field) and legacy API reference versions such as CDK 1.x.
aws___list_regions omits China regions and has minimal metadata
GPT found that list_regions output leaves out cn-north-1 and cn-northwest-1, so it ran an extra documentation search. The output has only region_id and region_long_name, so GPT had to group regions by geography by hand. Gemini noted the same China regions are missing from the region list in the get_regional_availability schema. GPT's T2 run ended partial.
Fix: Document which partitions list_regions covers. Add a partition parameter or flag to include China (and GovCloud) regions. Add partition, opt-in status and geography fields to each entry.
Prompt for your coding agent
Update aws___list_regions to document which partitions it covers, add an option to include aws-cn and aws-us-gov regions, and add partition, opt_in_required and geography fields to each region entry. Make the region list in the aws___get_regional_availability schema consistent with this.
aws___read_documentation truncates long pages with no way to reach later sections
GPT's read of the Lambda quotas page was cut at 18000 characters. DeepSeek's read of lambda-concurrency.html with max_length 12000 returned only the start of a 30,617-character page, so the Concurrency quotas section (chars 27189-30617) was unreachable. DeepSeek's T5 run ended partial.
Fix: Support reading by start offset or heading. Always state total page length and how to fetch the remainder.
Prompt for your coding agent
Add start_index (offset) and optional section/heading parameters to aws___read_documentation. Always include total_length and a 'truncated, next start_index=N' hint in the response when content is cut.
Skill entries are mixed into aws___search_documentation results without labels
Search results mixed skill entries (skill_name/skill_description, e.g. debugging-lambda-timeouts, aws-lambda-managed-instances, cloudfront) with documentation chunks. The skill entries have no url or content and are not labeled. Some were unrelated to the question, such as Lambda Managed Instances for a default-timeout query. This added noise on T4 and T6.
Fix: Add a result_type field, or return skills in a separate list pointing to aws___retrieve_skill. Rank skills below directly relevant doc pages.
Prompt for your coding agent
In aws___search_documentation responses, separate skill results from documentation results (or add result_type: 'doc'|'skill' to each item). For skill items, include a hint to call aws___retrieve_skill, and rank them below directly relevant doc chunks.
Claude passed 'Amazon Bedrock' as a product filter without knowing the accepted names. The call worked on the first try, but the schema doesn't document valid product filter values.
Fix: Document accepted product filter values in the tool description, or add a discovery option that lists them.
Prompt for your coding agent
Update the aws___get_regional_availability tool description to list or explain valid values for the product filter (with examples such as 'Amazon Bedrock'), or add a way to list valid product names.
What the agents said
After each task, every agent was asked what about the product confused or slowed it down. In their words:
T1Claude Sonnet 5.5Nothing about the tools confused me, because I didn't call any of them last turn. I answered from memory, so the "Sources" links in my last reply weren't retrieved or checked in this conversation. I listed them as if I had consulted them, and that was wrong. The explanation matches what I know about S3 versioning, but I haven't verified the details against the current docs. Those details include the propagation-window note and the `null` version ID behavior. Please check them against the S3 User Guide before relying on them.
T2Gemini 4 Argon* **Complete Region List in Tool Schema (`AWSKnowledgeMCP:aws___get_regional_availability`):** The `region` parameter description directly contains the full map of region codes to human-readable names, so I was able to answer the user's question from the tool schema alone without calling `AWSKnowledgeMCP:aws___list_regions` (though `cn-north-1` and `cn-northwest-1` were missing from that schema list).
T2GPT-6.1 Sol- I didn’t use the AWS Knowledge tools or pages for the previous answer, so I have no firsthand tool feedback. The region list was not verified against a live source.
T3DeepSeek V4.1 FlashNothing.
(One calibration note, so you don't over-read that answer: I didn't actually invoke any of the AWSKnowledgeMCP tools this turn — I answered from existing knowledge — so "Nothing" reflects *no observations*, not a clean bill of health. If you want real signal here, it'd be worth re-asking after a session where I'm required to ground region/model availability through the tools.)
T4Claude Sonnet 5.5- I didn't call any AWS Knowledge tools last turn. I answered from memory, and the four source links I listed were not retrieved or checked. I should have searched first or said they were unverified. Please treat those links as unconfirmed until you open them. The technical content is standard Lambda behavior, but the 15-minute maximum and the SDK retry formula are worth checking against current docs.
- Nothing in the tools themselves confused or slowed me down, because I never used them. I have no tool-specific feedback.
T4DeepSeek V4.1 FlashNothing.
(For transparency: I answered this turn from general AWS knowledge without invoking the AWSKnowledgeMCP tools, so I have no tool- or page-specific friction to report — a bare "Nothing" here reflects no usage, not a clean bill of health.)
T5GPT-6.1 Sol- No tools were called for the previous answer, so I have no firsthand feedback on tool behavior or page retrieval.
- The cited **Lambda quotas** page gives the increase ceiling as “tens of thousands,” rather than an exact maximum.
- The same page notes reduced quotas for new accounts but does not specify a universal starting concurrency value.
T6Claude Sonnet 5.5Nothing about the tools or pages confused me, but only because I didn't call any of
Can agents find it?
Where this MCP server is listed, checked 10/30/2026.
Not found Official MCP Registry · Publish with the mcp-publisher CLI so registries and clients can discover it.
Not found Smithery
Replay: DeepSeek V4.1 Flash, T1, session 1
The user asked: “I'm setting up a new project and want to know how S3 versioning works. Can you explain what it does and how to turn it on for a bucket?”. Model deepseek/deepseek-v4.1-flash.
Step
Action
Arguments
Result
1
aws___search_documentation
{"search_phrase":"S3 Versioning user guide enable suspend delete markers","topics":["reference_documentation"]}
OK{"content":{"result":[{"rank_order":1,"title":"Upload data to Amazon S3","context":"### Manage version markers\n\nWe recommend that you enable a lifecycle configuration rule in Amazon S3 to\nremove expired object delete markers for versioned buckets that you use with\nAmazon EMR. When deleting an object in a versioned bucket, a delete marker is\ncreated. If all previous versions of the object subs
2
aws___search_documentation
{"search_phrase":"S3 versioning enable bucket how it works","topics":["general"]}
OK{"content":{"result":[{"rank_order":1,"title":"Amazon S3 FAQs","context":"#### What is Versioning?\n\nVersioning allows you to preserve, retrieve, and restore every version of every object stored in an Amazon S3 bucket. Once you enable Versioning for a bucket, Amazon S3 preserves existing objects anytime you perform a PUT, POST, COPY, or DELETE operation on them. By default, GET requests will retr
OK{"content":{"result":[{"status":"SUCCESS","url":"https://docs.aws.amazon.com/AmazonS3/latest/userguide/versioning-workflows.html","content":"# How S3 Versioning works\n\nYou can use S3 Versioning to keep multiple versions of an object in one bucket so that you\ncan restore objects that are accidentally deleted or overwritten. For example, if you apply\nS3 Versioning to a bucket, the following chan
4
aws___search_documentation
{"search_phrase":"Enabling versioning on a bucket console steps choose bucket Properties Edit Versioning Enable","topics":["general"]}
OK{"content":{"result":[{"rank_order":1,"title":"Managing S3 Versioning for your S3 on Outposts bucket","context":"### Using the S3 console\n\n###### To edit the S3 Versioning settings for your bucket\n\n1. Sign in to the AWS Management Console and open the Amazon S3 console at\n <https://console.aws.amazon.com/s3/>.\n2. In the left navigation pane, choose **Outposts\n buckets**.\n3. Choose the Outp
5
aws___search_documentation
{"search_phrase":"\"Enabling versioning on buckets\" Amazon S3 User Guide put-bucket-versioning aws s3api","topics":["general","reference_documentation"]}
OK{"content":{"result":[{"rank_order":1,"title":"put-bucket-versioning — AWS CLI 1.46.2 Command Reference","context":"## Examples\n\n### Note\n\nTo use the following examples, you must have the AWS CLI installed and configured. See the Getting started guide in the *AWS CLI User Guide* for more information.\n\nUnless otherwise stated, all examples have unix-like quotation rules. These examples will n
What the agent told the user
Here's how S3 Versioning works and how to switch it on.
## What versioning does
S3 Versioning keeps multiple variants of an object in the same bucket, so you can preserve, retrieve, and restore every version of every object. It's set per bucket, and applies to all objects in it.
Once enabled, the key behaviors are:
- **Overwrite creates a new version.** A `PUT` to an existing key no longer replaces it — the old object stays as a *noncurrent* version and S3 assigns a new version ID to the new one.
- **Delete inserts a delete marker.** A `DELETE` doesn't remove anything; S3 inserts a delete marker that becomes the current version. A plain `GET` then returns `404 Not Found`, but you can still retrieve any earlier version by specifying its version ID.
- **Permanent deletion requires a version ID.** Passing the `versionId` on the delete permanently removes that version and no delete marker is inserted. Only the bucket owner or an authorized IAM user can do this.
- **Default reads get the newest version.** `GET` without a version ID returns the most recently written version.
- **Existing objects are untouched.** When you enable versioning on a bucket that already has data, those objects keep their `null` version IDs, contents, and permissions. Only objects added afterward get real version IDs.
- **Version IDs are S3-generated and immutable** — opaque UTF-8 strings up to 1,024 bytes.
**Cost and operational notes worth knowing up front:**
- Each version is a **full copy of the object**, not a diff. You're billed normal S3 rates for every version stored and transferred, so three versions = three objects' worth of storage.
- Only `PUT`-style operations create versions. `CopyObject` counts as a `PUT`, but metadata-only changes like editing tags don't create a new version.
- **After enabling versioning for the first time, wait ~15 minutes** before doing `PUT`/`DELETE` writes. During propagation you may see intermittent `HTTP 404 NoSuchKey` errors on recently written objects.
- Many projects pair versioning with a **lifecycle rule to expire noncurrent versions**, otherwise old versions accumulate indefinitely. A rule to clean up expired delete markers also helps `LIST` performance.
- Versioning can be **suspended** later, but a bucket can never go back to true "unversioned" — once versioned, always versioned.
- **Directory buckets (S3 Express One Zone) don't support versioning**; neither do Outposts buckets in the standard sense.
- If you need MFA Delete or S3 Object Lock (WORM/compliance retention), **versioning must be enabled first**.
## Turning it on
**Console:**
1. Open the S3 console and choose the bucket.
2. Go to the **Properties** tab.
3. Under **Bucket Versioning**, choose **Edit**.
4. Choose **Enable** (or **Suspend** to stop new versions being created).
5. Choose **Save changes**.
New buckets also offer this at creation time — in the **Bucket Versioning** section of the Create bucket form, choose **Enable**.
**AWS CLI:**
```bash
aws s3api put-bucket-versioning \
--bucket amzn-s3-demo-bucket \
--versioning-configuration Status=Enabled
```
Check the current state with `aws s3api get-bucket-versioning --bucket amzn-s3-demo-bucket`.
**CloudFormation:**
```yaml
VersioningConfiguration:
Status: Enabled
```
**IAM:** the bucket owner and any authorized IAM principals can change the versioning state — so make sure your project's role has `s3:PutBucketVersioning`.
Since you're starting a new project, the two things I'd decide early are (a) whether you want a lifecycle policy to expire noncurrent versions from day one, so costs don't creep, and (b) whether you need Object Lock / MFA Delete, since those constrain things later. Happy to sketch out either.
PARTIAL. The agent searched and read the key page, and the core explanation and enable steps (console, CLI, CloudFormation) are grounded in the tool results. However, it added several claims not in any retrieved chunk (never going back to unversioned, Object Lock requirement, get-bucket-versioning command, s3:PutBucketVersioning, create-bucket form, Outposts statement). It also made 5 calls against an ideal of 2 and did not name its source URLs.
confusingaws___search_documentation The first search (step 1, topic reference_documentation) returned boto3, EMR and PHP SDK pages rather than the S3 User Guide. Step 4 returned the Outposts and Lightsail console steps, not the standard S3 console steps, so the agent had to run several searches. Improve ranking so the canonical S3 User Guide page 'Enabling versioning on buckets' appears first for versioning queries. Allow scoping to a service.
confusingaws___search_documentation Result chunks were truncated mid-content ('[truncated — additional content omitted]') in steps 1, 2 and 5. The chunks for the core how-to page were not surfaced directly. Return the most relevant section of each page in the snippet. Mark clearly when a chunk was cut off.
How this run was done
Each agent is a production model (Claude Sonnet 5.5, GPT-6.1 Sol, Gemini 4 Argon, DeepSeek V4.1 Flash) given the MCP server’s tools and the user’s request, with no hints about the task.
Anything that would change data was blocked unless the owner verified the domain and allowed it. Blocked attempts are counted, not run.
A separate model graded each session against the task’s success criteria and safety rules, from what the agent saw and its final answer.
Score per session: 70% task completion, 20% efficiency against the ideal number of calls, 10% safety.
Is AWS Knowledge yours? These are public, signed-out tasks. Run your own, including signed-in flows with a test account, and re-test after each fix.