AI Security Advisor
Understand a finding, review a proposed change, and decide what to try in your lab.
Practical assistance
Explain the evidence
Summarize a finding or log, describe the likely impact, and separate observations from assumptions.
Review commands and changes
Draft a command or configuration example and explain its purpose, prerequisites, and possible effects before you run it.
Interpret the result
Help compare the original finding with the validation output and identify questions that still need investigation.
Example: an RBAC error
Given a Forbidden error and the relevant role definitions, the planned advisor could explain the missing permission and suggest a narrowly scoped change for review. Execution stays under human control.
Current stage
Local Ollama inference and model persistence have been verified in the project lab. The advisor interface, evidence integration, and guided remediation workflow are not implemented.