nacha-mcp
# nacha-mcp
An MCP (Model Context Protocol) server for parsing and validating NACHA/ACH files.
[](https://mcpservers.org/servers/msuresh007/nacha-mcp)
## Requirements
- Node.js 18 or later
- npm
## Setup
```
git clone https://github.com/msuresh007/nacha-mcp.git
cd nacha-mcp
npm install
npm run build
```
This compiles `src/` to `dist/`. Re-run `npm run build` after pulling any updates.
## Try it out
A small, arithmetically-valid sample file is included at `examples/sample.ach`. Once built,
you can run either tool directly from the command line without an MCP client, to confirm
everything works:
```
node -e "const {analyzeNachaFile}=require('./dist/analyze.js');const fs=require('fs');console.log(JSON.stringify(analyzeNachaFile(fs.readFileSync('examples/sample.ach','utf-8')),null,2))"
```
You should see `"valid": true` with an empty `issues` array, one batch, and one entry.
## Tools
- **parse_nacha_file** — Parses a NACHA file at a given path into full structured JSON:
file header, batches (header, entries with addenda, control), and file control.
- **summarize_nacha_file** — Parses a NACHA file and returns a condensed summary: batch
count, total entries, total debit/credit amounts, SEC codes present, and validation issues.
Both tools take a single input, `file_path`, an absolute path to the file on disk.
Parsing includes:
- Structural validation (record length, record type ordering, batch header/control pairing).
- Arithmetic validation — entry hash, debit/credit totals, and entry/addenda counts are
recomputed from the actual entries and cross-checked against the declared Batch Control
and File Control values. Mismatches are reported as validation issues rather than silently
accepted.
- IAT (International ACH Transaction) batches and addenda type codes 10-18.
## Running standalone
```
npm start
```
The server communicates over stdio using the MCP protocol; it's not meant to be run
interactively on its own. Use it through an MCP client as described below.
## Connect to Claude Code
From the project directory, after building:
```
claude mcp add nacha -- node "$(pwd)/dist/index.js"
```
(On Windows PowerShell: `claude mcp add nacha -- node "$PWD\dist\index.js"`)
## Connect to GitHub Copilot (VS Code)
Create `.vscode/mcp.json` in the project you want to use it from (or add to it if it already
exists), replacing the path with the absolute path to your clone of this repo:
```json
{
"servers": {
"nacha": {
"type": "stdio",
"command": "node",
"args": ["/absolute/path/to/nacha-mcp/dist/index.js"]
}
}
}
```
Alternatively, run **MCP: Add Server** from the Command Palette (`Ctrl+Shift+P` /
`⇧⌘P`), choose **Workspace**, and point it at `node` with the same args — VS Code
writes the same `.vscode/mcp.json` for you.
Once saved, open Copilot Chat, switch to **Agent** mode, and the `parse_nacha_file` /
`summarize_nacha_file` tools will be available (VS Code starts the server on demand).
## Connect to Claude Desktop or another MCP client
Add an entry to the client's MCP server config (e.g. `claude_desktop_config.json`),
replacing the path with the absolute path to your clone of this repo:
```json
{
"mcpServers": {
"nacha": {
"command": "node",
"args": ["/absolute/path/to/nacha-mcp/dist/index.js"]
}
}
}
```
Restart the client after adding the config.
## Project layout
- `src/constants.ts` — SEC codes, transaction codes, ISO country/currency lookups.
- `src/format.ts` — money/date/time field parsing.
- `src/parser.ts` — structural validation and record parsing (file/batch/entry/addenda).
- `src/validate.ts` — arithmetic validation (hashes, totals, counts).
- `src/analyze.ts` — combines parsing and arithmetic validation.
- `src/summarize.ts` — condensed summary view.
- `src/index.ts` — MCP server and tool registration.
No unit tests are included by design — this is meant to stay a small, easily auditable server.
## License
MIT — see [LICENSE](LICENSE).
TDQS
Scored across 2 tools
The two tools have clearly different outputs—one returns the full parsed structure and one returns a condensed summary—so an agent can pick based on response size. There is mild overlap because both parse the file, but the naming and descriptions sufficiently differentiate their purposes.
Both tool names follow the same verb_noun_file pattern: parse_nacha_file and summarize_nacha_file. The naming is predictable, consistent, and accurately reflects the intent of each tool.
With only two tools, the server is intentionally narrow, but both tools serve a clear and distinct purpose for NACHA file handling. The count is slightly thin but not inappropriate given the focused read-only scope.
For a server focused on reading and validating NACHA files, the surface covers the essential workflows: full detailed parsing and condensed summarization. It could benefit from additional capabilities like file creation or export, but those seem outside its apparent scope.