A1 ratingsAWSAmazon Bedrock AgentCore

Amazon Bedrock AgentCore

AWS · platform · Any industry
A
AGrade
70Score
±1495% range
≈#69Rank of 1134
0%From fleets
FR-CAT-00838Catalog no.

The six pillars

Each bar is the score; the bracket is its 95% range.

Can you see what it costs?

Cloud billing export (AWS Cost Explorer API (GetCostAndUsage) / Cost and Usage Reports) · prices published
63±38
Desk research 76 · counts for 25% · A cost API you can call scores highest; an admin console or cloud billing export next; a card statement or emailed invoice is all some products leave you. Published prices add to it.

Can you see how much it's used?

Amazon CloudWatch metrics (observability billed at CloudWatch rates for AgentCore)
86±27
Desk research 100 · counts for 15% · A usage API (tokens, seats, credits, conversations) a buyer can read.

Can you tie cost to one agent?

By project
64±27
Desk research 70 · counts for 15% · Exact (one key or project per agent) beats by-project, by-workspace or by-user.

Can an agent on it report out?

SDK · Agent can call AWS SDK (boto3) for Cost Explorer and CloudWatch with an IAM role.
64±27
Desk research 70 · counts for 15% · A remote MCP connector scores highest, then a local MCP server, an SDK, an API or a workflow step; copy-and-paste lowest. Any open standard counts, not just FleetRegistry.

Can you control spend and access?

Spend cap: Alerts only · Per-agent keys: Scoped · Audit log: By API · Revoke: By API · SSO: SSO + SCIM
69±38
Desk research 88 · counts for 20% · Hard spend caps (30%), scoped per-agent keys (25%), an audit log you can read by API (20%), revoking by API (15%), SSO with SCIM (10%).

Does it work on small plans?

Small plans get the data
86±27
Desk research 100 · counts for 10% · Whether the plan a person or small business buys gets the cost and usage data at all.

Where the evidence comes from

Independent fleets0%Desk research62%Neutral starting point38%

Once three independent fleets report on a pillar, desk research counts for at most a third of it. FleetRegistry's own fleets never count.

Controls on record

Spend capAlerts only
Per-agent keysScoped
Audit logBy API
RevokeBy API
SSOSSO + SCIM
PricesPublished

Price changes: AWS Customer Agreement s.3.1: at least 30 days' prior notice for fee increases on services in use

There is no dollar-based hard cap. The AgentCore harness caps iterations, time and tokens per invocation (carried over from the prior check), and AWS Budgets actions can apply a Deny IAM policy or SCP but don't guarantee charges stop. Bedrock API keys are tied to IAM principals and long-term keys can be deactivated via the API. CloudTrail and IAM Identity Center ratings carry over from the prior check.

History

11-12 A 7011-13 A 7011-14 A 7011-15 A 7011-16 A 7011-17 A 7011-18 A 7011-19 A 70
First rated 2026-10-02 at 69 (B). Now 70. Published grades are sealed daily into the transparency log (N4.4). Grade held since 2026-10-16.

Every fact, with its rung and source

Each fact stands on its own source and its own rung of the evidence ladder (Rule N2.1). Verified Observed Declared Inferred. The weakest fact in a pillar sets its weight.

FactValueRungCheckedSources
How a buyer sees cost
Cost
Cloud billing export Declared
medium confidence
2026-11-15
2 earlier
[1] Correct
Prices published
Cost
Yes Inferred 2026-11-15
3 earlier
[1][2][3] Correct
Usage API
Usage
Amazon CloudWatch metrics (observability billed at CloudWatch rates for AgentCore) Declared
medium confidence
2026-11-15
2 earlier
[1] Correct
Cost per agent
Per agent
By project Declared
medium confidence
2026-11-15
2 earlier
[1] Correct
How an agent reports out
Reports out
SDK Declared
medium confidence
2026-11-15
2 earlier
[1] Correct
Per-agent keys
Controls
Scoped Inferred 2026-11-15
3 earlier
[1][2][3] Correct
Audit log
Controls
By API Inferred 2026-11-15
4 earlier
[1][2][3] Correct
Revoke access
Controls
By API Inferred 2026-11-15
3 earlier
[1][2][3] Correct
Spend cap
Controls
Alerts only Inferred 2026-11-15
3 earlier
[1][2][3] Correct
SSO
Controls
SSO + SCIM Inferred 2026-11-15
3 earlier
[1][2][3] Correct
Small plans get the data
Small plans
Yes Declared
medium confidence
2026-11-15
2 earlier
[1] Correct

Prices · Runtime/Browser/Code Interpreter $0.0895/vCPU-hr + $0.00945/GB-hr (Runtime v2 $0.1276/vCPU-hr + $0.0169/GB-hr); Gateway $0.005/1k invocations; short-term Memory $1.00/GB ingest, $0.20/GB retrieval, $0.10/GB-mo; long-term Memory $0.25-$0.75/1k records/mo; Policy $0.000025/request; plus model inference

Notes · AWS-billed, with prices read from the AWS AgentCore pricing page. Short-term memory is consumption-priced per GB (moved from per-event pricing in Oct 2026). Runtime v2 also lists a committed-baseline rate ($0.0997/vCPU-hr, $0.0132/GB-hr), and Gateway Search, Identity, Policy token and Evaluations charges are listed separately. Observability is billed at CloudWatch rates.

Scored with the rubric for Platforms and APIs (Rule N7.1). Each pillar blends a neutral 50, desk facts weighted by rung, and independent fleets. A1 is 85+ with fleet proof, A 70, B 55, C 40, D 25, E below. Part N of the Rules

Corrections

1 awaiting review. Accepted corrections are published in the corrections register with an FR-C number.

Run agents on Amazon Bedrock AgentCore?

Give each one a permanent number, a named person who answers for it, and a record of what it can touch and what it costs. Free.