AI assistants and search decision
Amazon Q Developer
A tiny MCP-like proxy and skill runner can be built, but AWS's global infrastructure and deep, maintained integration with hundreds of services are durable advantages that a single developer or small team cannot reproduce, so keep paying for the managed toolkit if you need enterprise-grade coverage and controls.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need — the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship.
What a replacement has to do
- Provide agents with an authenticated endpoint that proxies AWS API actions, serves curated 'skills' (scripts/recipes), executes sandboxed Python scripts, and logs/audits agent activity.
What it still won’t have
- AWS's global scale, availability zones, and managed infrastructure
- Deep integration with 240+ AWS services and automatic updates
- Enterprise controls and built-in CloudWatch/IAM guardrails at AWS scale
- Managed MCP Server SLA and automatic plugin updates for third-party agents
What remains hard
- Infrastructure at scale
The AWS Cloud spans 123 Availability Zones within 39 Geographic Regions, with announced plans for 7 more Availability Zones and 2 more AWS Regions in the Kingdom of Saudi Arabia, and Chile.
- Integration maintenance
Modernize faster and scale more efficiently with an unmatched portfolio of over 240 comprehensive services
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 4 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal self-hosted MCP-like server in Python (FastAPI) on a single EC2 instance that: 1) exposes an authenticated HTTP endpoint to accept MCP-style requests and map them to AWS SDK (boto3) calls; 2) serves a small skill repository (YAML/JSON) loaded from S3 or local git and fetchable at runtime; 3) runs submitted Python skill scripts in a sandboxed interpreter with no network/filesystem access (use subprocess + user namespaces or a container sandbox like Firecracker or gVisor); 4) enforces IAM-scoped credentials per-agent and validates requested actions against a rules file; 5) writes audit logs and basic metrics to S3/CloudWatch; Out of scope: replicating multi-region availability, managed auto-scaling, or deep integration with 240+ AWS services. Include error handling, retries for AWS API calls, authentication tests, and unit tests for proxy, sandbox, and policy enforcement.
How we checked
How the score was reached
- Pay verdict base20
- An open-source build was found+5
- 5 cited sources+3
- Hard moats found in the evidence-3
- Evidence score25
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time — so the same evidence always produces the same number.
How scoring works →Cited sources · 5
Every page the run actually retrieved.
- official productAWS home
- official pricingAWS Pricing
- official docsAgent Toolkit for AWS
- open sourcearc53/DocsGPT
- open sourcehangwin/mcp-chrome
Integrity checks
What held up, and what did not.
