detect_schema_driftLABSCO SUMMARY
Detection only — the fixing is the paid edition.
It answers the question that matters before a deploy: between two versions of a subject, what changed, and which of it breaks downstream consumers. What the name promises beyond that — AST rewriting, regenerated test payloads, automated PRs — belongs to the paid Enterprise edition, not this repository.
What it is
A Python server against a Confluent Schema Registry: it lists the registered subjects and diffs two versions of one, classifying each field change as breaking or backward-compatible. This is the free Community Edition.
What you get
- Every registered subject in the Schema Registry listed
- Two versions of a subject diffed — added, removed and modified fields
- Each change classified as breaking or backward-compatible
- A drift report naming the fields, their types and their defaults
Requirements
Python, and pip install -r requirements.txt; the client runs python server.py as a stdio server. The registry URL, API key and API secret are passed as call arguments rather than environment variables, so the credentials travel with each request. Code rewriting, test-payload generation and PR creation are not in this edition.
Setup effort
Build from source — git clone https://github.com/rohitdas-ai/kafka-schema-drift-auto-fixer-free
USE CASES
01
Checking whether a schema change is safe to ship
Diff the two versions and read which fields were added, removed or retyped, with the breaking ones marked.
02
Finding the subject you need
List every registered subject in the registry rather than guessing the name.
03
Explaining a consumer failure after the fact
Diff the versions either side of the deploy and see what the consumers were reading against.
04
Reviewing a schema pull request
Run the drift check against the registered versions as part of the review, before anyone merges.
2 tools
Listed in project docs. Your client may show a different list.
list_subjectsBEFORE YOU INSTALL
What to know first
Tools you get
2 tools named in the project's documentation.
Requires
Python with pip
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
git clone https://github.com/rohitdas-ai/kafka-schema-drift-auto-fixer-free
cd kafka-schema-drift-auto-fixer-free
pip install -r requirements.txtMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"kafka-schema-drift-community": {
"command": "python",
"args": [
"<YOUR_PATH_TO_KAFKA_SCHEMA_DRIFT_AUTO>/server.py"
]
}
}
}<YOUR_PATH_TO_KAFKA_SCHEMA_DRIFT_AUTO> is the absolute path of your clone — server.py is at its root. The README's example writes this path as /path/to/community-edition/server.py.
