Glossary
Plain-English definitions for the TENSA ecosystem.
A quick reference for the language used across TENSA Engineering, NeuroCore, Argus ACLI, Argus Lab, controlled AI systems, and Linux diagnostics.
How to use this page
Short definitions first. Deeper reading second.
This page is meant to support the rest of the site. Terms on project pages and Knowledge Base articles can link here when a reader needs a quick human definition without leaving the larger flow of the page.
Each definition stays short, explains the term in plain English, gives TENSA context, and links back to the best deeper page when one exists.
Term index
Browse by category.
Use the links below as a quick map. Each term jumps to a short definition lower on the page.
TENSA ecosystem
Controlled AI
Diagnostics
System awareness
Training
Definitions
Human-readable term definitions.
These are intentionally short. The goal is to explain the term clearly, connect it to the TENSA ecosystem, and point readers toward the deeper page when they want more context.
AI Operations
Human definition: The discipline of making AI-assisted work repeatable, grounded, reviewable, and continuous.
TENSA context: AI Operations includes context loading, source-of-truth documents, build logs, resume prompts, review loops, and closeout workflows.
Read more: AI Operations.
Argus ACLI
Human definition: The read-only Linux diagnostics interface built on NeuroCore.
TENSA context: Argus ACLI turns Linux system signals into structured findings, severity, recommendations, and raw evidence.
Read more: Argus ACLI.
Argus Lab
Human definition: A real-Linux troubleshooting training and validation environment built around controlled faults and real symptoms.
TENSA context: Argus Lab is where future learners practice troubleshooting and where Argus diagnostic behavior can be tested against known failures.
Read more: Argus Lab.
Authority boundary
Human definition: The line between what an AI model may reason about and what it is allowed to control.
TENSA context: Authority boundaries keep model explanation separate from governed execution paths.
Read more: Controlled AI Systems.
Control plane
Human definition: The governance layer that decides what actions are allowed before tools or operations run.
TENSA context: NeuroCore’s control plane helps separate AI reasoning from operational authority.
Read more: NeuroCore Architecture.
Controlled AI system
Human definition: An AI-assisted system where intelligence and authority are deliberately separated.
TENSA context: The model may reason and explain, but system interaction must pass through governed paths.
Read more: Controlled AI Systems.
Diagnostic intelligence
Human definition: The ability to turn system evidence into useful troubleshooting insight.
TENSA context: Argus builds diagnostic intelligence from real service state, logs, memory, disk, network, and process evidence.
Read more: Linux Diagnostics.
Evidence-first troubleshooting
Human definition: Starting an investigation from real system evidence instead of guesses.
TENSA context: Argus and Argus Lab both emphasize finding, preserving, and reasoning from evidence.
Read more: Linux Diagnostics.
Governed execution
Human definition: Tool or system actions pass through policy and control boundaries instead of being launched directly by the model.
TENSA context: Governed execution is how NeuroCore keeps AI useful near real systems without giving it unchecked control.
Read more: Safe Tool Interaction.
Linux diagnostics
Human definition: Inspecting Linux system state to understand what is happening and why.
TENSA context: Argus ACLI focuses on services, logs, processes, memory, disk, network state, and raw command evidence.
Read more: Linux Diagnostics.
Local-first AI
Human definition: AI designed to work close to the user, the system, and the evidence by default.
TENSA context: Local-first design keeps sensitive system context closer to the owner and closer to the real environment.
Read more: Local-First AI.
MCP-based tool
Human definition: A tool connected to an AI application through the Model Context Protocol, an open standard for exposing external data, tools, and workflows.
TENSA context: An MCP-based connection can give an AI model structured access to approved external systems, but the permissions and authority boundaries still need to be defined and reviewed.
Read more: Safe Tool Interaction.
Model authority
Human definition: What an AI model is allowed to decide or control.
TENSA context: TENSA limits model authority so explanation and reasoning do not become unchecked system control.
Read more: Controlled AI Systems.
NeuroCore
Human definition: The governed local-first AI platform behind TENSA Engineering systems.
TENSA context: NeuroCore provides the foundation for controlled tool interaction, structured evidence, system awareness, continuity, and AI-assisted explanation.
Read more: NeuroCore.
Operational continuity
Human definition: Preserving enough project or system context to keep moving without constantly restarting from memory.
TENSA context: TENSA uses source-of-truth documents, build logs, state records, and context loading to preserve continuity.
Read more: Persistent AI Memory.
Persistent AI memory
Human definition: Stored context that helps an AI system remember useful project information across sessions.
TENSA context: TENSA treats persistent memory as useful only when it stays grounded, reviewable, and connected to source-of-truth material.
Read more: Persistent AI Memory.
Raw evidence
Human definition: The original system signal behind a diagnostic finding.
TENSA context: Raw evidence can include command output, service status, journal logs, process data, ports, disk usage, and direct system signals.
Read more: Linux Diagnostics.
Read-only diagnostics
Human definition: Inspecting system state without changing the system.
TENSA context: Argus ACLI starts read-only so users can gather evidence and understand the problem before considering repair.
Read more: Argus ACLI.
Safe tool interaction
Human definition: Letting AI work near tools without giving the model unchecked control over those tools.
TENSA context: NeuroCore structures and governs tool interaction so the model does not become a raw command launcher.
Read more: Safe Tool Interaction.
Severity
Human definition: A label that helps describe how important or risky a diagnostic finding appears to be.
TENSA context: Argus uses severity to help users distinguish normal state, warnings, degraded conditions, and critical signals.
Read more: Linux Diagnostics.
Source-of-truth
Human definition: The trusted reference used to decide what is actually true about a project, system, or page.
TENSA context: Source-of-truth documents prevent public pages and AI-assisted work from drifting into memory, assumptions, or stale summaries.
Read more: AI Operations.
Structured findings
Human definition: Diagnostic results organized into clear fields such as status, severity, evidence, recommendation, and explanation.
TENSA context: Structured findings make Linux diagnostics easier for humans to read and easier for AI systems to explain.
Read more: Argus ACLI.
System awareness
Human definition: Grounded information about the real environment an AI-assisted system is helping explain.
TENSA context: NeuroCore’s direction is built around structured awareness from real evidence, controlled tools, context, and memory.
Read more: NeuroCore Architecture.
Telemetry
Human definition: Data collected from a system so its state or behavior can be understood.
TENSA context: Diagnostic telemetry can include service state, logs, process data, network information, and other system evidence.
Read more: Linux Diagnostics.
Tool boundary
Human definition: The separation between AI reasoning and the tools that interact with a real system.
TENSA context: Tool boundaries let the model request, interpret, or explain information without directly running unrestricted commands.
Read more: Safe Tool Interaction.
Troubleshooting training
Human definition: Structured practice in identifying, investigating, fixing, and verifying real technical problems.
TENSA context: Argus Lab is designed to support troubleshooting training through controlled real-Linux scenarios and mentor-style guidance.
Read more: Troubleshooting Training.
Verification
Human definition: Checking that a conclusion, fix, page update, or system state is actually true.
TENSA context: Verification matters across diagnostics, documentation, AI-assisted writing, website updates, lab scenarios, and project continuity.
Read more: AI Operations.