mcpField
C
cloudsession

cloudsession

Local
github.com/CloudCorpRecordsCloud & Infra0Updated 1w ago
Repository

About cloudsession

cloudsession is an MCP server for managing local AWS session context and execution plans, aimed at workflows that need lightweight state without calling AWS APIs. Its in-memory sessions and ordered plans are advanced explicitly by the caller.

Key features

  1. 01Start, inspect, and end sessions
  2. 02Set AWS profile and region
  3. 03List supported regions
  4. 04Create and inspect ordered plans
  5. 05Advance plan steps explicitly
  6. 06In-memory state for server lifetime

Use cases

  1. 01Tracking a local AWS profile and region context
  2. 02Organizing an ordered sequence of execution steps
  3. 03Managing multi-step state through one MCP server process

MCP support

Exposes nine tools for session and plan state to MCP clients. The tools operate on local in-memory state and do not call AWS APIs.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "cloudcorprecords-cloudsession": {
      "command": "npx",
      "args": [
        "-y",
        "@reneturcios/cloudsession"
      ]
    }
  }
}
Run commandnpx -y @reneturcios/cloudsession

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@reneturcios/cloudsession0.1.0
ocighcr.io/cloudcorprecords/cloudsession:0.1.0

Tools

9

Saved tool list · updated 1w ago

ParameterTypeRequiredDescription
regionstringNoAWS region for the new session, such as "us-east-1". Must be a known region; defaults to "us-east-1" when omitted.
profilestringNoAWS profile name to associate with the new session. Defaults to "default" when omitted.

Details

From the project README

cloudsession

Local AWS session and plan state as MCP tools. cloudsession is a clean-room rebuild of the session/plan meta core of the AWS core MCP server, under a new name: real local state logic, fresh code, no copied branding. No tool in this subset calls an AWS API — a session is a local context carrying a profile name and a region, and a plan is an ordered list of steps advanced explicitly by the caller.

Tools (core subset, 9)

start_session, get_session_info, end_session, set_region, set_profile, list_regions, create_plan, advance_plan_step, get_plan.

Sessions start with profile default and region us-east-1 unless overridden, get deterministic IDs (session-1, plan-1, ...), and validate regions against a curated static region list (29 regions). Ending a session blocks further mutation while keeping it readable. Plans track each step as pending/completed; advancing the final step completes the plan, and advancing a completed plan is an error. State is in-memory for the life of the server process — nothing persists across restarts.

Not in this subset: AWS API proxy/call tools, credential and SSO login flows, documentation search helpers, prompt templates, and any tool that calls live AWS service APIs.

Setup

No credentials or environment variables are required.

CLI

npx @reneturcios/cloudsession list_regions '{}'
npx @reneturcios/cloudsession start_session '{"region":"eu-west-1"}'
npx @reneturcios/cloudsession mcp        # stdio MCP server

Each tool name is a subcommand taking one JSON object argument. Note that each CLI invocation starts with a fresh store, so multi-step session/plan flows belong in the MCP server, which keeps one store for its lifetime.

MCP client config

{
  "mcpServers": {
    "cloudsession": {
      "command": "npx",
      "args": ["-y", "@reneturcios/cloudsession", "mcp"]
    }
  }
}

Docker

docker run -i ghcr.io/cloudcorprecords/cloudsession

Development

npm install
npm run build
npm test

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.