Run the Docker Compose demo to upload a sample bank statement and see the planned audit flow in action.
Study the Go gateway's in-memory redaction approach as an example of stripping PII before an AI processing step.
Use the project's gRPC contract between the Go gateway and Python service as a reference for a similar multi-language architecture.
Extend the Python auditor service to add real CSV parsing or subscription detection logic beyond the current demo stub.
| uiansol/vigil-auditor | 42wim/fabio | 42wim/go-xmpp | |
|---|---|---|---|
| Stars | 0 | — | — |
| Language | Go | Go | Go |
| Last pushed | — | 2018-02-04 | 2020-01-24 |
| Maintenance | — | Dormant | Dormant |
| Setup difficulty | hard | moderate | moderate |
| Complexity | 4/5 | 3/5 | 3/5 |
| Audience | developer | ops devops | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires Go, Python 3.12+, Node.js, and Docker Compose to bring up the gateway, AI service, database, and frontend together.
Vigil is a project that aims to help people find and cancel forgotten subscriptions and creeping charges by analyzing their bank statements. Instead of connecting to a bank account through a third party aggregator service, a user uploads a raw PDF or CSV bank statement, and Vigil looks for hidden subscriptions, bills that quietly increase over time, and free trials that turned into paid charges without the user noticing. It also includes a planned cancellation feature meant to walk a user through the steps needed to cancel an unwanted service, including direct links where possible. The project's stated priority is privacy. Uploaded statements are streamed in memory rather than written to disk, and the system strips out sensitive details like account numbers before any data is sent to an AI model, so only merchant names are used for further processing. The README states that raw uploads are not saved in this first version of the project. Under the hood, Vigil is split into a few pieces working together. A Go based gateway service handles file uploads, session cookies, and streams progress updates back to the browser as the audit runs. A separate Python service handles the actual parsing and analysis, matching similar merchant names together and detecting billing creep, with a plan to use a locally run AI model in a later stage of development. A Next.js and React frontend gives users a dashboard to upload files and watch the audit progress live, and PostgreSQL stores session and audit data. This is an early stage, in progress project. The current slice of work supports uploading a file and seeing a staged demo progress bar, while real statement parsing is planned for a later slice. Running it requires Go, Python, Node.js, and Docker with Docker Compose to bring up all the services together, or a hybrid setup running the backend in Docker and the frontend directly on your machine. The README does not state a license.
Vigil is an in-progress tool that scans uploaded bank statements to find hidden subscriptions and billing creep without sending raw data to a bank aggregator.
Mainly Go. The stack also includes Go, Python, Next.js.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.