MCP STDIO proxy for Amazon Bedrock AgentCore Runtime
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-agentcore-proxy","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# Amazon Bedrock AgentCore MCP Proxy ## Overview This repository addresses two limitations in Amazon Bedrock AgentCore when running MCP servers: 1. **IAM Authentication**: AgentCore requires OAuth 2.0 or Cognito for authentication through the gateway.[^gateway-inbound][^cognito-auth] Developer workstations typically use IAM credentials via Identity Center instead. This repository packages a local stdio proxy that signs requests with SigV4, allowing direct invocation of the AgentCore runtime API[^runtime-how] without configuring OIDC providers or VPC access. 2. **Stateful Bidirectional MCP**: AgentCore's native MCP protocol support is stateless. MCP sampling (server-to-client LLM requests) and elicitation (schema-driven data collection) require persistent sessions. This repository provides an HTTP-to-STDIO bridge that maintains a subprocess across invocations, enabling bidirectional MCP flows within AgentCore's serverless infrastructure. The local proxy handles IAM authentication and presents a standard MCP stdio interface to clients. Two deployment models are provided for the AgentCore runtime: - **Stateless**: Direct MCP server for simple request/response tools - **Stateful**: HTTP bridge maintaining a persistent MCP subprocess for sampling and elicitation workflows [](https://pypi.org/project/mcp-agentcore-proxy/) [](https://pypi.org/project/mcp-agentcore-proxy/) [](https://github.com/alessandrobologna/agentcore-mcp-proxy/actions) [](https://opensource.org/licenses/MIT) ## Quick Start If you already have an MCP server deployed on Amazon AgentCore Runtime…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.