| Query | A natural language question sent to AI for analysis. Consumes from your org's monthly pool. |
| Live Session | Interactive diagnostic session with one agent via a three-tab modal (Query, Shell, Files). |
| Multi-Agent Session | Diagnostic session spanning multiple agents with parallel query distribution and cross-correlation. |
| Cross-Correlation | Cloud-side analysis comparing findings across agents for common patterns and shared root causes. |
| Alert Rule | Configurable condition(s) that trigger notifications when agent metrics meet specified thresholds. |
| Notification Channel | Delivery endpoint for alerts (Email, Slack, Teams, Webhook). Reusable across rules. |
| Agent Seat | Billing unit for managed agents. Each managed agent = 1 seat. Desktop agents are free. |
| Registration Token | A secure etd_... string created by org admins for authenticating agent installations. Stored as a SHA-256 hash; plaintext shown only once at creation. Scoped to one organization, revocable, with optional max-agent limits and expiry dates. |
| Admin-Purchased Subscription | A query subscription bought by an organization admin on behalf of a team member. Billed to the org, manageable by admins. Contributes to the org's query pool and agent seat count like a self-purchased subscription. |
| Data Retention | Configurable duration that cloud-stored data (events, metrics, correlations, sessions) is kept before automatic purge. Free tier: 14 days. Paid add-on tiers extend to 90, 365, or 730 days. Billed per managed agent per month. |
| BYOK | Bring Your Own Key — free tier using your own Anthropic API key for unlimited queries. |
| Data Explorer | Visual query builder for custom analysis of fleet telemetry data. Supports multiple data sources, time-series bucketing, grouping, filtering, and chart visualization. Accessible from the Reports sub-navigation. |
| Historical AI Analysis | Conversational AI interface for asking natural language questions about fleet data. Powered by Claude. Translates questions into database queries, streams narrative insights with inline charts, and supports follow-up chains for multi-step exploration. Uses 1 AI query per analysis. |
| Smart Reports | Natural language query interface on the Reports page. Translates plain English questions into instant chart visualizations with data tables. Supports follow-up analysis with cross-source correlation: the system runs 2–5 correlation queries across health metrics, events, and agents, then streams a Claude-powered narrative analysis with root cause identification and recommendations. |
| Workspace | Your team's tenant on ET Ducky — the outer container that everything else lives inside. One workspace owns one URL (a subdomain on etducky.com) and holds one or more Clerk organizations. Workspace members automatically receive membership in every organization in the workspace. Provisioned on first sign-up; renamable from Settings → Workspace; previous slug remains active as a grace-period alias after a rename. |
| Subdomain | The DNS hostname (e.g. acme in acme.etducky.com) that routes to a workspace. The slug is editable through workspace settings; the workspace identity itself is permanent — renames preserve every binding. |
| Workspace Membership Cascade | Server-side fan-out that keeps workspace membership and Clerk org membership in lockstep. When a user joins a workspace, they are added to every organization in the workspace; when a new organization is created in the workspace, all existing workspace members are added to it. Implemented so “workspace member” and “has access to all orgs in the workspace” mean the same thing in practice. |
| Organization | Clerk-managed entity scoping all resources: agents, registration tokens, tags, alert rules, notification channels, queries, scripts, automations, tickets, integrations, and per-org members. Each organization belongs to exactly one workspace. Users can belong to multiple organizations within a workspace and switch between them via the org picker; dropdowns are filtered to only show organizations from the current workspace. Created from the Dashboard, Team, or Agent Setup pages. |
| Workspace Picker | Modal shown when a user signs in at etducky.com and belongs to two or more workspaces, listing each workspace and the user's role in it. Selecting one redirects to that workspace's URL. |
| Claim a Subdomain | First-sign-in flow for users who don't yet belong to a workspace. The user picks an available slug, ET Ducky provisions the workspace plus a default Clerk organization, and the user is redirected to the new URL. |
| SSE | Server-Sent Events — a lightweight HTTP-based protocol for real-time server-to-client streaming. Used for live sessions, multi-agent queries, agent heartbeat events, and remote desktop relay. |
| Agent Tag | Organization-scoped, color-coded label for agent categorization. Usable in alert rule conditions. |
| Remote Desktop | Browser-based screen sharing and remote control for agents. Uses DXGI screen capture, WebSocket relay, and Canvas rendering. |
| DXGI | DirectX Graphics Infrastructure — GPU-accelerated screen capture API used by the remote desktop feature for high-performance, low-CPU frame capture with dirty-region tracking. |
| Dirty Region | A rectangular area of the screen that changed since the last frame. Only dirty regions are encoded and transmitted, significantly reducing bandwidth. |
| Integrations | Third-party ticketing connections (Jira, ServiceNow) configured per organization. One active integration per org is used for ticket selection and push-to-ticket in Run Troubleshooting and for viewing integration tickets on the Tickets page. |
| Tickets (ET Ducky) | Native tickets stored in your organization. Created from the dashboard (e.g. Run Troubleshooting) or submitted by end users via the ET Ducky Support shortcut. Viewable on the Tickets page alongside integration tickets when viewing by agent. |
| ET Ducky Support | Desktop and Start Menu shortcut that opens a simple support-ticket form. Runs the agent with --support; no full desktop app. Submissions are sent to the organization that owns the agent and appear on the Tickets page. |
| OOB Gateway | A Raspberry Pi running the ETDucky.Gateway service. Connects to the ET Ducky cloud over WebSocket, scans the local network for AMT/DASH/IPMI-capable devices, and relays out-of-band commands from the dashboard to target hardware. |
| Out-of-Band (OOB) Management | Hardware-level remote management that operates independently of the host operating system. Enables power control, inventory, and diagnostics even when the OS is unresponsive or the machine is powered off. |
| Intel AMT | Intel Active Management Technology — firmware-level remote management available on Intel vPro business processors. Communicates over WS-Management on ports 16992 (plain) and 16993 (TLS). Requires a vPro-capable CPU and must be enabled in BIOS/MEBx. |
| DASH | Desktop and mobile Architecture for System Hardware — AMD's equivalent to Intel AMT, available on AMD PRO processors with compatible Broadcom or Realtek enterprise NICs. Communicates over WS-Management on port 623. |
| IPMI | Intelligent Platform Management Interface — server BMC (Baseboard Management Controller) standard for out-of-band server management. ET Ducky supports IPMI identify and reachability probes on port 623. |
| Gateway Token | An etd_gateway_-prefixed bearer token issued to a single Raspberry Pi gateway. Provisioned from the token creation dialog by selecting the Gateway token type. Distinct from Windows agent tokens; cannot be used for agent API endpoints. |
| Command Security Engine | Server-side evaluation layer that checks all shell commands for hard-blocked patterns (malware droppers, RAT installers, credential theft tools, etc.) and risk-scores remaining commands before they are queued for agent execution. Commands scoring 60 or above are blocked. |
| Intel MEBx | Management Engine BIOS Extension — a firmware setup menu (typically accessed by pressing Ctrl+P during POST) used to enable and configure Intel AMT on vPro hardware. |
| Automation Rule | An org-scoped rule that combines a trigger (alert fired, agent enrolled/offline/online, cron schedule, or manual), an action (run script, webhook, notify, create ticket), and a target scope (triggering agent, all agents, or tagged agents). Rules execute automatically when their trigger condition is met. |
| Org Script | A named PowerShell or CMD script stored at the organization level. Shared across Live Sessions, Fleet Tools, and Automations. Organized by category (Diagnostics, Maintenance, Security, Inventory, Compliance, Imported, Custom). System scripts are read-only; custom scripts are editable by org admins. |
| Automation Scheduler | Background service that evaluates cron-based automation rules every 60 seconds. When a rule’s next scheduled run time has passed, it executes the rule and recalculates the next run from the cron expression. All times are UTC. |