FanDesk

Glossary

Key terms used throughout FanDesk and its documentation.

A

Action Tagging Automatic AI classification of emails by required action. Tags are: NEEDS_ACTION (requires your response), ACTION_LATER (has a future deadline), AWAITING_RESPONSE (you sent an email and are waiting), and NO_ACTION (no response required — newsletters, automated notifications).

Active Hours Your configured working hours. Used for daily digest timing, quiet hours enforcement, and notification scheduling. Set in Settings > Profile.

AI Employee A named, persistent, role-specific AI worker built on FanDesk's knowledge layer. Unlike DeskMate (which answers questions), AI Employees handle entire workflows autonomously and accumulate institutional memory over time. First AI Employee: Riyan (External Relations). Coming soon.

Artifact A rich, interactive document created by DeskMate or via the MCP create_artifact tool. Artifacts can contain charts, metrics tables, visualizations, and dynamic components — not just static text.

Auto-Assignment Automatic assignment logic for unassigned work items. For tasks: unassigned tasks in a project are assigned to the project lead after 24 hours. For incidents: a three-layer system using rules, response teams, and manual fallback routes incidents to the right responder.

B

Brain Capital The combined intellectual output of your team and AI, working in concert. FanDesk's core mission is amplifying collective brain capital — the unified knowledge, reasoning capacity, and context that makes your organization smarter over time. When data is fragmented across tools, brain capital leaks. When unified in FanDesk, it compounds.

Burndown Chart Sprint analytics visualization showing the rate at which remaining work decreases over a sprint. Compares the ideal line (linear decrease to zero) against the actual remaining task count.

C

Channel A persistent group chat space for team discussions. Public channels are open to anyone in the workspace. Private channels are invite-only.

Channel Group A folder-like organization unit for channels in the sidebar. Create groups like "Engineering", "Clients", or "Projects" to organize channels and reduce sidebar noise.

Custom Field User-defined metadata field on tasks beyond the built-in fields. Types: text, number, date, select, multi-select, checkbox, URL, email, phone, person, files, and relation.

D

Daily Digest Personalized email summary sent at the start of your active hours each day. Includes open incidents, tasks due today, upcoming calendar events, and an AI-curated reading list (factoid + articles based on your preferences).

DeskMate FanDesk's built-in AI assistant with access to 250+ MCP tools across all modules. Uses a 3-tier model system that auto-upgrades based on query complexity: L3 (fast, default) → L2 (smart) → L1 (reasoning). Enriches every response with context from your knowledge graph.

DM (Direct Message) Private one-on-one conversation between two users. Only the two participants can see the messages.

E

Entity A named concept in the knowledge graph — a person, company, project, product, technology, or decision that FanDesk has identified and indexed from your data.

Entity Extraction Automatic AI identification and indexing of entities (people, companies, projects, concepts, decisions) from your messages, emails, tasks, and documents. The raw material that populates the knowledge graph. Runs every 15 minutes.

G

Guest An external collaborator with limited, scoped access. Guests can only see channels they've been invited to, pages shared with them, and tasks assigned to them. Guests cannot access mail, calendar, drive, or the full team directory.

I

IMAP Internet Message Access Protocol — the email protocol used to connect non-Gmail accounts (Zoho, Outlook, Yahoo, custom mail servers) to FanDesk. IMAP-connected accounts support all the same email features as Gmail accounts, including AI classification, action tagging, and thread summarization.

Incident An unplanned event that disrupts or threatens to disrupt service. Has a severity level (Critical, High, Medium, Low), a category (Outage, Security, Performance, Bug, Infrastructure, Data), and a lifecycle (Triggered → Acknowledged → Investigating → Resolved → Postmortem).

K

Kanban A project board view with columns representing task statuses. Drag tasks between columns to change their status. Kanban is one of four task views (alongside List, Table, and Calendar).

Knowledge Graph FanDesk's automatically maintained network of entities and relationships extracted from all your data across every module. Powers semantic search, DeskMate context enrichment, and relationship discovery. Gets smarter and more connected the more your team uses the platform. Embeddings refresh every 10 minutes; entity extraction runs every 15 minutes; graph maintenance runs daily at 2:30 AM.

M

MCP (Model Context Protocol) An open standard backed by Anthropic and the Linux Foundation for connecting AI assistants to applications via a standardized tool interface. FanDesk exposes 250+ tools via MCP, allowing Claude Code, Claude Desktop, Cursor, and any MCP-compatible client to manage your entire FanDesk workspace programmatically.

MTTA (Mean Time to Acknowledge) Key incident response metric: the average time from when an incident is created (Triggered) to when a responder first acknowledges it. Target: under 15 minutes for Critical incidents.

MTTR (Mean Time to Resolution) Key incident response metric: the average time from when an incident is created (Triggered) to when it is marked Resolved. Varies by severity but should trend downward over time.

P

Pinned Message A message bookmarked at the top of a channel for easy reference. Pinned messages are visible to all channel members via the pin icon in the channel header.

Postmortem A structured document created after a resolved incident. Contains: summary, root cause, impact assessment, timeline, what went well, what went wrong, and action items (which become trackable FanDesk tasks). Published postmortems are searchable via the knowledge graph.

Pulse FanDesk's native video and audio meeting system, embedded directly in Chat. Supports live recording, real-time transcription in 99 languages, participant management (mute, remove, admit from waiting room), virtual backgrounds, and AI-generated post-meeting summaries (TL;DR, decisions, action items). Completed Pulses are searchable and linkable to calendar events and email threads.

S

Semantic Search AI-powered search that finds results based on meaning and intent, not just exact keyword matches. Powered by Voyage AI vector embeddings (1024-dimensional vectors). FanDesk uses semantic search across all modules — tasks, pages, emails, messages, contacts — via the knowledge graph.

Sprint A fixed-duration iteration (typically 1-2 weeks) in which a team commits to completing a specific set of tasks. FanDesk tracks sprint velocity, burndown, and completion rate to help teams improve predictability over time.

Split Inbox Email organization system that categorizes incoming emails into tabs: Important, Team, Updates, Newsletters, and Other. Classification uses a combination of Gmail's categories, pattern matching, and AI analysis. Categories are customizable via rules.

Status Your current availability indicator, visible to all teammates: Online (green dot), Away (yellow dot), Do Not Disturb (red dot), Offline (gray dot). Set manually or automatically based on activity.

T

Task Key A unique, searchable identifier for every task in FanDesk. Format: PROJECT_KEY-NUMBER (e.g., BACK-42, MOBILE-7, INC-15). Searchable from the command palette and linkable in documents and messages.

Thread A focused, nested conversation attached to a specific message. Threads keep detailed discussion out of the main channel flow and are accessible via the "Reply in thread" action on any message.

Tier System DeskMate's three-level AI model escalation system: L3 (fast — default for all queries), L2 (smart — auto-upgrades after 3+ tool calls), L1 (reasoning — for the hardest analytical problems). Upgrades happen automatically; you never need to choose a tier.

V

Vector Embeddings High-dimensional mathematical representations of text that capture semantic meaning. FanDesk uses Voyage AI (voyage-4 model, 1024-dimensional vectors) to embed all content across every module. Embeddings power semantic search and knowledge graph similarity matching.

Velocity Sprint analytics metric measuring how many tasks (or story points, if configured) a team completes per sprint. Tracking velocity over multiple sprints helps teams forecast capacity for future sprints and identify improvement trends.


You've reached the end of the FanDesk documentation. Need help? Ask DeskMate in Chat, or reach out at hello@fandesk.ai.

Need help? Contact us at hello@fandesk.ai