DeciFlow
An AI real-time log classification decision-layer tool for operations scenarios. It inserts a structured classification step between raw logs and human judgment, sorting logs into five categories: Normal, Warning, Error, Security, and Other. It returns category probabilities and confidence scores, enabling operations staff to prioritize suspicious events.
Timeline (last 90 days)
DeciFlow inserts a structured classification step between raw logs and human judgment, sorting logs into one of five categories: Normal, Warning, Error, Security, or Other
DeciFlow's classification output includes category probabilities and confidence information
DeciFlow supports three log input methods: Paste mode, Batch mode, and Realtime mode
DeciFlow supports multi-file and wildcard patterns, so logs do not need to be manually merged beforehand
DeciFlow nodes can monitor logs locally while sending inference requests to a remote shared Laya server, and the log source machine does not need a GPU or a PyTorch installation
DeciFlow supports two backends: Laya is the default option, supports self-hosting, and can run on CPU or CUDA; Jev is an alternative hosted decision API
DeciFlow treats AI classification results as reference rather than truth, the system can flag predictions that need human review, operators can view the original events and record corrections, and results can be exported as JSON
DeciFlow integrates Python Actions; an Action receives the JSON data of a successful decision and executes custom Python code, with an independent Python environment and dependency management
DeciFlow currently only uses the choice decision type, that is, selecting one from preset categories
The DeciFlow author reveals plans to add in the future a score type for evaluation and scoring, and a noul type for yes/no probability judgments
1 more timeline events
All Facts (11)
DeciFlow uses containerized deployment, the repository contains Docker instructions for the DeciFlow UI and the Laya inference server, and the browser interface runs on port 4000 by default
50%UnverifiedDeciFlow's classification output includes category probabilities and confidence information
50%UnverifiedDeciFlow supports three log input methods: Paste mode, Batch mode, and Realtime mode
50%UnverifiedDeciFlow supports multi-file and wildcard patterns, so logs do not need to be manually merged beforehand
50%UnverifiedDeciFlow nodes can monitor logs locally while sending inference requests to a remote shared Laya server, and the log source machine does not need a GPU or a PyTorch installation
50%UnverifiedDeciFlow supports two backends: Laya is the default option, supports self-hosting, and can run on CPU or CUDA; Jev is an alternative hosted decision API
50%UnverifiedDeciFlow treats AI classification results as reference rather than truth, the system can flag predictions that need human review, operators can view the original events and record corrections, and results can be exported as JSON
50%UnverifiedDeciFlow integrates Python Actions; an Action receives the JSON data of a successful decision and executes custom Python code, with an independent Python environment and dependency management
50%UnverifiedDeciFlow currently only uses the choice decision type, that is, selecting one from preset categories
50%UnverifiedDeciFlow inserts a structured classification step between raw logs and human judgment, sorting logs into one of five categories: Normal, Warning, Error, Security, or Other
50%UnverifiedThe DeciFlow author reveals plans to add in the future a score type for evaluation and scoring, and a noul type for yes/no probability judgments
50%