ABRA Flexi MCP Server
Connects to ABRA Flexi over its REST API to read companies, evidence, partners, products, balances, and overdue items, while supporting guarded write flows with validation and audit logging.
README
ABRA Flexi MCP Server
Local stdio MCP server for ABRA Flexi / FlexiBee accounting workflows. It is intended for agent clients such as Codex, Claude Desktop, or any MCP-compatible local runtime that can launch a Node process.
This is the standalone MCP project. The separate remote ChatGPT App project is here: vvitovec/abra-flexi-chatgpt-app.
What It Does
- connects to ABRA Flexi over the official REST API
- reads companies, evidence metadata, records, partners, products, balances, overdue items, and accounting summaries
- prepares accountant-focused document drafts
- supports guarded write flows with validation, dry-run style checks, confirmations, and an audit log
- keeps Flexi credentials out of prompts and MCP tool arguments
Safety Model
The server is configured through flexi.config.json profiles. Each profile controls:
- Flexi base URL and company slug
- test or production mode
- default response format
- which env vars contain credentials
- read, dry-run, and write evidence allowlists
- whether writes require confirmation
Writes are not free-form passthrough calls. Tool handlers build known payloads, validate them, enforce evidence permissions, and record request audit files under .flexi-harness/logs.
Setup
npm install
cp .env.example .env
Edit .env with a dedicated ABRA Flexi REST API user:
FLEXI_PROD_USERNAME=api-user
FLEXI_PROD_PASSWORD=replace-me
Then edit flexi.config.json:
{
"defaultProfile": "prod",
"profiles": {
"prod": {
"baseUrl": "https://example.flexibee.eu",
"company": "example_company_s_r_o_",
"usernameEnv": "FLEXI_PROD_USERNAME",
"passwordEnv": "FLEXI_PROD_PASSWORD"
}
}
}
Run Locally
npm run dev
For a production-style build:
npm run build
npm start
Codex MCP Config
After building, add a server entry like this to your Codex config:
[mcp_servers.abra_flexi]
command = "node"
args = ["/absolute/path/to/abra-flexi-mcp-server/dist/index.js"]
enabled = true
The server loads .env and flexi.config.json from the project directory.
Main Tool Areas
- company and evidence discovery
- flexible read-only evidence queries
- partner and product search
- invoice, payable, receivable, bank, cash, and internal document workflows
- overdue and saldo summaries
- accountant-first draft creation and guarded posting
- latest-error explanation from the local audit log
Verification
npm run check
npm test
npm run build
The tests use mocked HTTP servers and fixture-style responses. They do not need real ABRA Flexi credentials.
Related Docs
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.