$ loading_
Detect ML drift, diff MLflow registries, and perform HMAC-gated model rollbacks.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "Mendrift" MCP server from askskill: Run: claude mcp add 'io-github-suneel190700-mendrift-mcp' -- npx -y mendrift-mcp
Use Mendrift to detect drift in the current production model and summarize any abnormal metrics.
Returns drift detection results and highlights metrics or data changes that may have shifted.
Compare two MLflow model registry versions and list the key differences.
Produces a summary of differences between the two registry versions for model change review.
After HMAC validation, perform a model rollback and state the target version.
Returns the rollback result and indicates that the action is protected by HMAC gating.
DevOps or ML engineering teams can use it after deployment to detect drift and catch changes in data distribution or model behavior. This helps identify production issues earlier.
When teams iterate on model versions in MLflow, they can use it to diff registry entries during release review. This makes key changes between versions easier to inspect.
If a newly deployed model causes issues, teams can use an HMAC-gated flow to roll it back. It suits environments that require controlled and secure rollback operations.
Based on the description, it is an MCP server for machine learning drift detection, MLflow registry diffing, and HMAC-gated model rollback.
The description explicitly mentions MLflow registry diffing, so at least that capability involves the MLflow registry. For exact dependencies and configuration, see the source repository.
The description says model rollback is HMAC-gated, which means the rollback flow requires an HMAC validation mechanism. For key management and exact validation details, see the source repository.
Monitor production ML drift and degradation with alerts and retraining guidance.
Expose Django REST Framework APIs as secure MCP tools for AI agents.
Understand codebases, detect conventions, and retain decisions across sessions.
Audit project portfolios against GitHub conversationally to detect drift and anomalies.
Lets AI assistants manage LLM inference, backends, and VRAM via MCP.
Analyze disk images with AI through MCP for fast forensic investigation.