The Data Recovery Taxonomy & Framework gives the Enterprise Recovery Orchestrator app its underlying vocabulary and process model. The taxonomy classifies every recovery record along four simultaneous axes — Recovery Domain (Disk, Database, Cloud, Email, Cyber Artifact), Outcome (Recovered/Partial/Failed), Confidence Band (High/Medium/Low), and Tooling — so any record can be described precisely regardless of what kind of system it came from. The framework, called KDRF, lays out the five-phase process every case moves through: Identify & Triage, Acquire & Preserve, Analyze, Recover & Reconstruct, and Validate, Score & Report. A cross-cutting Decision Layer runs across the final three phases, acting as an "Agentic AI Recovery Advisor" that turns analyzer output into a confidence score, recovery plan, time estimate, and tool recommendation. Five formulas quantify the whole thing — Recovery Probability, Confidence Level, Estimated Time, record-level confidence bands, and domain-weighted success rates — making the taxonomy and framework not just descriptive but computable.
LaunchThe Enterprise Recovery Orchestrator is a Streamlit app that simulates recovery operations across five domains — Disk, Database, Cloud, Email, and Cyber Artifact Recovery — by generating synthetic telemetry records with randomized status, confidence score, size, and tool used, weighted per domain. It persists every run and its individual records to a local SQLite database (`knet_recovery.db`), enabling a History tab that lists past runs and can clear them. A rule-based "Agentic AI Recovery Advisor" (`compute_ai_recommendation()`) turns the generated records into a Recovery Probability, Confidence rating, Estimated Time, and a prioritized list of suggested actions and tools, explicitly designed as a stand-in for a future live LLM call. Results are visualized via Plotly (pie, stacked bar, histogram) with a native-chart fallback, and the Export tab offers CSV, JSON, TXT, PDF, and DOCX downloads, with PDF/DOCX gated behind `reportlab`/`python-docx` availability checks. Overall it's the operational simulation engine that pairs with the Taxonomy & Framework app: this one runs and scores synthetic recoveries, the other documents the vocabulary and math behind them.
Launch