# Libra AI > Libra is an AI agent platform that connects your work tools, understands your context, and handles multi-step work across the systems you already use. Every URL below is a public, crawlable page. The authenticated app and user-shared content are excluded on purpose; see /robots.txt. ## Product - [Libra: the AI agent platform built on your company’s work](https://trylibra.ai/): Libra is an AI agent platform that connects your work tools, understands your context, and handles multi-step work across the systems you already use. - [Libra for Enterprise: one brain for everything your company knows](https://trylibra.ai/enterprise): Libra WorkBase gives enterprises a connected context layer for knowledge, reasoning, governed execution, permissions, and deployment across company systems. - [AI Agents — Libra](https://trylibra.ai/product/agents): Libra Agents take on multi-step work across your connected tools and company knowledge, executing tasks end to end with control when approval matters. - [Integrations — Libra](https://trylibra.ai/product/integrations): Connect Libra to the tools your company already uses. Bring email, meetings, documents, CRM, projects, and databases into one context and work layer. - [Libra Knowledge Base: from information to answers](https://trylibra.ai/product/knowledge-base): Libra Knowledge Base brings company documents, conversations, and information together so people and agents can ask questions and work from trusted sources. - [Libra Meeting Assistant: AI meeting intelligence for your team](https://trylibra.ai/product/meeting-assistant): Libra Meeting Assistant helps you prepare for meetings, capture what matters, and turn decisions and action items into work across your connected tools. - [Libra Email Assistant: replies that already know the answer](https://trylibra.ai/product/email-assistant): Libra Email Assistant helps you understand, write, and act across email using the context from your connected company tools and workflows. - [AI Business Intelligence — Libra](https://trylibra.ai/product/business-intelligence): Libra Business Intelligence lets teams ask questions in plain language, query live data, and turn results into charts, tables, and dashboards. - [AI Presentation Maker — Libra](https://trylibra.ai/product/slides): Libra is an AI presentation maker that turns a brief into structured slides, presentation-ready copy, visuals, layouts, and a finished deck. - [AI Workflows: Ready-to-Deploy Automation for Teams — Libra](https://trylibra.ai/product/workflows): Prebuilt AI workflows your team can deploy in minutes: follow-up chasers, issue triage, feedback digests. Automation that runs on your company's context. - [Libra Pricing: start personal, give your company a shared brain](https://trylibra.ai/pricing): Explore Libra pricing for self-deployed, workplace, and enterprise plans, with options for hosted teams, company knowledge, agents, connectors, and governance. - [Contact Libra Sales: set up your team’s company brain](https://trylibra.ai/contact-sales): Book a Libra demo to see WorkBase run a real task across your tools, workflows, permissions, and company context. - [Download Libra for Mac](https://trylibra.ai/download): Download Libra Neo for Mac and use AI from your desktop to dictate, rewrite, capture notes, and take action without leaving your workflow. - [Libra at the cafe: work sessions on us](https://trylibra.ai/cafe): Claim a voucher, bring your laptop, and try Libra on real work over coffee. Limited seats at each session. ## Solutions - [Libra WorkBase for Sales: turn calls into CRM updates](https://trylibra.ai/solutions/sales): Libra WorkBase for Sales turns calls, emails, and CRM activity into updates, follow-ups, research, and next steps so sales teams can focus on deals. - [Libra WorkBase for Customer Support: resolve tickets faster](https://trylibra.ai/solutions/customer-support): Libra WorkBase for Customer Support connects tickets, customer history, and company knowledge to triage issues, draft replies, and move escalations forward. - [Libra WorkBase for Engineering & Product: decisions become work](https://trylibra.ai/solutions/engineering-product): Libra WorkBase for Engineering & Product turns conversations, specs, tickets, and decisions into structured work across Jira, Linear, Notion, and more. - [Libra WorkBase for Recruitment: keep the hiring process moving](https://trylibra.ai/solutions/recruitment): Libra WorkBase for Recruitment helps teams research candidates, prepare interviews, coordinate calendars, update the ATS, and keep hiring moving. - [Libra WorkBase for Marketing: from brief to campaign](https://trylibra.ai/solutions/marketing): Libra WorkBase for Marketing connects research, briefs, assets, campaigns, and reporting to move work from planning through distribution. - [Libra WorkBase for Founders & Executives: decide faster](https://trylibra.ai/solutions/founders-executives): Libra WorkBase for Founders & Executives connects company updates, metrics, meetings, and decisions into briefs, prep, and follow-through. ## Compare - [Glean Alternatives & Competitors: Libra vs Glean](https://trylibra.ai/compare/glean): Glean is built to find and understand what your company already knows. Libra starts from the same context but carries it into execution — agents that finish the… - [Genspark Alternatives: Why Teams Pick Libra](https://trylibra.ai/compare/genspark): Genspark turns a prompt into a finished artifact — a deck, a sheet, a video. Libra starts from your company's context and lands the work in the systems you already… - [Kore.ai Alternatives: Why Teams Pick Libra](https://trylibra.ai/compare/kore-ai): Kore.ai gives enterprises a platform to build and govern agents on. Libra gives your team a workspace that already works, with the control built in — agents doing… - [Dust Alternatives & Competitors: Libra vs Dust](https://trylibra.ai/compare/dust): Dust gives technical teams a fast environment to build custom agents. Libra ships the context layer and the execution on top, so employees get finished work… - [Claude Cowork Alternatives: Libra vs Claude Cowork](https://trylibra.ai/compare/claude-cowork): Claude Cowork is a desktop AI coworker for one person's computer work. Libra runs on shared company context across the whole team's tools, with approvals and… - [Manus Alternatives & Competitors: Libra vs Manus](https://trylibra.ai/compare/manus): Manus is a general-purpose autonomous agent that builds its context inside each task. Libra runs on persistent company knowledge that carries across tools and… - [ChatGPT Enterprise Alternatives: Libra vs ChatGPT Enterprise](https://trylibra.ai/compare/chatgpt-enterprise): ChatGPT Enterprise is the fastest way to put a capable general-purpose model in front of a whole company. Libra is built around a persistent knowledge layer that… - [Microsoft Copilot Alternatives: Libra vs Copilot](https://trylibra.ai/compare/microsoft-copilot): Copilot is deepest where your company already lives in Microsoft 365. Libra is built for stacks that also span Gmail, Slack, Salesforce and Notion. Compare… ## Blog - [Blog | Libra AI](https://trylibra.ai/blog): Read the Libra AI blog for practical guides, research, and perspectives on AI agents, enterprise AI, workflows, integrations, and the future of work. - [AI Agent Observability: What to Instrument Before Agents Touch Production](https://trylibra.ai/blog/ai-agent-observability): What to instrument when an agent runs multi-step work on your systems — the signals that matter, the failure modes request-level monitoring misses, and where… - [What the Meta–Manus Unwind Means for Anyone Running AI Agents](https://trylibra.ai/blog/meta-manus-unwind-what-it-means-for-ai-agents): Meta bought Manus for $2B, Beijing blocked it, and the deal was unwound. What it means if your AI agent vendor changes hands and the three questions to ask before… - [Free Knowledge Base Software: What You Actually Get for $0](https://trylibra.ai/blog/free-knowledge-base-software): Compare free knowledge base software, including Onyx, Libra, Notion, and Confluence. Understand free-plan limits, self-hosting costs, and when to upgrade. - [How to Build an Internal Knowledge Base: A Step-by-Step Guide](https://trylibra.ai/blog/how-to-build-an-internal-knowledge-base): Learn how to build an internal knowledge base in six steps, from identifying recurring questions to assigning content owners, connecting tools, and measuring… - [Internal Knowledge Base Software: 8 Tools Compared (2026)](https://trylibra.ai/blog/internal-knowledge-base-software): Compare 8 internal knowledge base software tools, including Notion, Slite, Onyx, and Libra. Explore pricing, self-hosting, AI features, and how to choose. - [Agentic AI Platforms: What They Are, How to Evaluate Them, and Where They Break](https://trylibra.ai/blog/agentic-ai-platform): A complete guide to agentic AI platforms, how they differ from traditional AI, the use cases that work, evaluation criteria, deployment models, honest limitations… - [AI Agents for Customer Support: How They Work, What to Automate, and How to Choose](https://trylibra.ai/blog/ai-agent-for-customer-support): Learn how AI agents for customer support work, what to automate first, how to choose a platform, and how to manage implementation, security, and performance. - [Context Engineering: The Discipline That Decides Whether Your AI Works](https://trylibra.ai/blog/context-engineering): Learn what context engineering is, how it differs from prompt engineering, its four core patterns, common failure modes, and how enterprises can apply it. - [AI Agent Workflows: How to Structure Work an Agent Can Actually Finish](https://trylibra.ai/blog/ai-agent-workflows-how-to-structure-work-an-agent-can-actually-finish): Learn how to structure AI agent workflows with clear goals, scoped tools, human checkpoints, explicit failure handling, and output contracts. Explore common… - [Customer Service Knowledge Base: What to Build and What to Skip](https://trylibra.ai/blog/customer-service-knowledge-base-what-to-build-and-what-to-skip): Learn how to build a customer service knowledge base that answers real customer questions, stays accurate, and reduces support tickets through better content and… - [Enterprise Knowledge Management Software: A Buyer's Guide for 2026](https://trylibra.ai/blog/enterprise-knowledge-management-software): Compare enterprise knowledge management software by category, security, connectors, governance, pricing, and deployment to choose the right approach for your… - [Support Knowledge Base Software: How to Choose in 2026 | Libra AI](https://trylibra.ai/blog/support-knowledge-base-software): Compare support knowledge base software for customer self-service and internal agent support, including Zendesk, Intercom, Freshdesk, Document360, Guru, Notion… - [What Are AI Agents? A Plain-English Explanation | Libra AI](https://trylibra.ai/blog/what-are-ai-agents): An AI agent takes a goal, decides what steps to take, uses tools, and checks whether it worked. Learn what agents are made of, how they work, where they succeed… - [What Is an AI Workspace? The Category, Explained | Libra AI](https://trylibra.ai/blog/ai-workspace): An AI workspace gives AI persistent access to company context, tools, and workflows. Learn what defines the category and how to evaluate one. - [Enterprise AI Search: How It Works and How to Evaluate It](https://trylibra.ai/blog/enterprise-ai-search): Learn how enterprise AI search works, how it differs from traditional enterprise search, and what to evaluate around accuracy, permissions, freshness, and citations. - [Information Silos: Why They Form, and Why Reorganising Your Tools Won't Fix Them](https://trylibra.ai/blog/information-silos): Information silos form when knowledge gets trapped across teams, tools, and people. Learn why they happen and what actually helps break them down. - [Tribal Knowledge: What It Is and Why It Never Gets Written Down](https://trylibra.ai/blog/tribal-knowledge): Tribal knowledge is the expertise companies rely on but rarely document. Learn why it stays undocumented, what it costs, and how AI can help surface it. - [Agentic AI vs AI Agents: What's Actually Different | Libra AI](https://trylibra.ai/blog/agentic-ai-vs-ai-agents): Agentic AI and AI agents aren't the same. Learn how they differ, how much agency businesses actually need, and what to consider when deploying AI agents. - [What an AI Email Assistant Actually Does (and Where It Should Stop)](https://trylibra.ai/blog/ai-email-assistant): Discover what an AI email assistant does, how it uses context to draft replies, and where human approval should remain in the email workflow. - [AI Workflow Automation: How It Differs From Rule-Based Automation](https://trylibra.ai/blog/ai-workflow-automation): Learn how AI workflow automation differs from traditional workflow automation software, when to use each, and how AI workflows handle unexpected cases. - [Enterprise Search Software: What It Does and What It Cannot Do](https://trylibra.ai/blog/enterprise-search-software-what-it-does-and-what-it-cannot-do): Learn how enterprise search software connects company data, retrieves relevant information, and where search falls short compared with AI answer layers. - [Internal Knowledge Base: What It Is and Why Most Go Stale](https://trylibra.ai/blog/internal-knowledge-base): Learn what an internal knowledge base is, what it should contain, why knowledge bases go stale, and how teams can capture knowledge from everyday work. - [AI Agent Governance: A Practical Framework for Enterprise Rollout](https://trylibra.ai/blog/ai-agent-governance-a-practical-framework-for-enterprise-rollout): AI agent governance defines the permissions, approvals, oversight and operating rules enterprises need to deploy agents safely at scale. - [AI Agent Security: The Real Risks and How Enterprise Buyers Should Evaluate Vendors](https://trylibra.ai/blog/ai-agent-security): AI agent security is about more than model safety. Enterprise buyers need permission-aware access, governed actions, approvals, auditability and deployment controls. - [Enterprise AI Agents: Identity, Accountability, and What It Takes to Run Them](https://trylibra.ai/blog/enterprise-ai-agents): Enterprise AI agents act on real systems, which makes identity and accountability the hard problem — not capability. Use cases, governance, integration, evaluation… - [AI Assistants: What They Are, How They Work & Use Cases](https://trylibra.ai/blog/ai-assistants-what-they-are-how-they-work-use-cases): What is an AI assistant? Learn how modern AI assistants work, how they differ from chatbots and AI agents, and what business tasks they can automate. - [Knowledge-Based Software: What It Is & How It Works | Libra AI](https://trylibra.ai/blog/knowledge-based-software-what-it-is-how-it-works): Learn what knowledge-based software is, how it works, how it differs from traditional software, and how businesses use AI to turn company knowledge into action. - [What Is Agentic Process Automation? How AI Agents Change Business Workflows](https://trylibra.ai/blog/agentic-process-automation): Learn what agentic process automation is, how AI agents differ from traditional automation, and where enterprises can use agentic workflows. - [What Is a Knowledge Base? AI Knowledge Management Guide](https://trylibra.ai/blog/what-is-a-knowledge-base): Learn what a knowledge base is, how knowledge management works, and how AI knowledge bases help teams find and use company information faster. - [AI Coworkers and Business Systems: From Data to Action | Libra AI](https://trylibra.ai/blog/ai-coworkers-and-business-systems-from-data-to-action): AI coworkers become useful when they can connect business systems and execute workflows across apps. Learn how cross-app AI changes enterprise automation. - [Best AI Knowledge Management Platforms in 2026 | Libra AI](https://trylibra.ai/blog/best-ai-knowledge-management-platforms-in-2026): Compare the best AI knowledge management platforms in 2026 by retrieval, verification, capture, pricing, deployment, and fit for your organization. - [Human + AI Coworker Workflows: How Teams Should Divide Work](https://trylibra.ai/blog/human-ai-coworker-workflows-how-teams-should-divide-work): AI coworkers shouldn't replace humans blindly. Learn how to divide work between people and AI using accountability, review, escalation, and workflow design. - [What Integrations Does an Enterprise AI Assistant Actually Need?](https://trylibra.ai/blog/what-integrations-does-an-enterprise-ai-assistant-actually-need): Not all enterprise AI integrations are created equal. Learn which systems an AI assistant needs, why connector depth matters, and what to evaluate beyond connector… - [Enterprise Search vs RAG vs AI Agents vs Company Brain | Libra AI](https://trylibra.ai/blog/enterprise-search-vs-rag-vs-ai-agents-vs-company-brain): Enterprise search, RAG, AI agents, and a company brain solve different problems. Learn how they fit together and which layer your enterprise AI needs. - [Permission-Aware AI: Access Control for Enterprise AI | Libra AI](https://trylibra.ai/blog/permission-aware-ai-access-control-for-enterprise-ai): Learn how permission-aware AI enforces access control at retrieval time, inherits source-system permissions, and prevents sensitive data exposure in enterprise AI. - [BYOC for Enterprise AI: Cloud vs On-Prem Explained | Libra AI](https://trylibra.ai/blog/byoc-enterprise-ai-deployment): BYOC runs the vendor's software in your own AWS, Azure or GCP account, your VPC, your keys, your logs. When it beats vendor cloud, and when it doesn't. - [What Is a Company Brain? | Libra AI](https://trylibra.ai/blog/what-is-a-company-brain): What is a company brain? Learn how a company brain connects knowledge, context, decisions, people, and business systems and how it differs from a knowledge base… - [Agentic AI Architecture: Core Components Explained | Libra AI](https://trylibra.ai/blog/agentic-ai-architecture-core-components-explained): Learn how agentic AI architecture works, from context and reasoning to planning, orchestration, tools, memory, agents, and human oversight. - [Agentic AI Companies to Know in 2026 | Libra AI](https://trylibra.ai/blog/agentic-ai-companies-to-know-in-2026): A category map of the agentic AI market in 2026, control planes, vertical agents, general-purpose agents, and self-hosted platforms, with what each type is… - [The Best AI Agent Platform in 2026: How to Actually Run the Evaluation](https://trylibra.ai/blog/the-best-ai-agent-platform-in-2026-how-to-actually-run-the-evaluation): The five criteria that decide it, a procurement checklist, and how to design a pilot that predicts production. - [Enterprise AI Platforms: 2026 Buyer’s Guide | Libra AI](https://trylibra.ai/blog/enterprise-ai-platforms-buyers-guide): Compare enterprise AI platforms in 2026. Learn what to evaluate across AI agents, knowledge, integrations, security, deployment, governance, and cost - [How to Build an AI Agent: A Step-by-Step Guide | Libra AI](https://trylibra.ai/blog/how-to-build-an-ai-agent): Building an agent is less about picking a model than designing the system around it, the goal, context, tools, permissions, memory, and escalation rules that… - [What Is AI Orchestration? A Clear Explanation | Libra AI](https://trylibra.ai/blog/what-is-ai-orchestration): Orchestration is the coordination layer that turns model capability into a controlled, multi-step system, the part that decides which tool runs, what state is… - [AI Agents for Marketing: Use Cases and Practical Applications](https://trylibra.ai/blog/ai-agents-for-marketing-use-cases-and-practical-applications): Explore AI marketing agents and how they automate content, SEO, campaign coordination, competitor monitoring, reporting, and personalization. - [What Are AI Sales Agents? Use Cases, Benefits & Examples](https://trylibra.ai/blog/ai-agents-for-sales-applications-and-benefits): Discover how AI sales agents automate lead research, outreach, CRM updates, scheduling, follow-ups, and more while helping sales teams save time. - [Cloud vs. Self-Hosted AI Agents: Key Differences Explained](https://trylibra.ai/blog/cloud-vs-self-hosted-ai-agents-a-framework-for-selecting-a-deployment-model): Compare cloud and self-hosted AI agents across data ownership, security, cost, control, and deployment. Use this framework to choose the right model. - [Rebuilding Libra’s memory as a system](https://trylibra.ai/blog/rebuilding-libras-memory-as-a-system): We opened up how Libra remembers, found parts that weren't doing anything, and redesigned it around one idea: memory is a system you design, not a feature you add. - [Autonomous AI Agents: How They Work, Architecture, and Enterprise Use Cases](https://trylibra.ai/blog/autonomous-ai-agents-how-they-work): What autonomous AI agents are, how the reason-act loop works, the architecture behind them, real enterprise use cases, and how to bound autonomy safely. - [Prithvi | Libra AI](https://trylibra.ai/blog/author/prithvi): Prithvi Bharadwaj writes about AI, enterprise software, and the changing nature of work. At Libra, he focuses on making complex ideas around AI agents, company… - [Mansi | Libra AI](https://trylibra.ai/blog/author/mansi): Mansi is a marketing professional at Libra AI, focused on building products and stories around how AI can fundamentally change the way people work. With a… ## Glossary - [The Libra Dictionary: an AI glossary for enterprise teams](https://trylibra.ai/glossary): Clear, one-sentence definitions for the concepts behind enterprise AI: context layers, agents, governed execution, permissions, memory and the rest. - [AI Agent — definition | Libra AI](https://trylibra.ai/glossary/ai-agent): An AI system that carries out a multi-step task across tools, rather than returning a single answer. - [AI Governance — definition | Libra AI](https://trylibra.ai/glossary/ai-governance): The permissions, approval rules and audit requirements that determine what an AI system may do. - [Ambient Agent — definition | Libra AI](https://trylibra.ai/glossary/ambient-agent): An agent that runs in the background on a signal rather than a prompt, surfacing work instead of waiting to be asked. - [Approval Workflow — definition | Libra AI](https://trylibra.ai/glossary/approval-workflow): A checkpoint that requires human sign-off before an AI agent completes a consequential action. - [Audit Trail — definition | Libra AI](https://trylibra.ai/glossary/audit-trail): The durable record of what an AI system did, in which system, on whose authority, and whether a person approved it. - [Business Intelligence (AI) — definition | Libra AI](https://trylibra.ai/glossary/ai-business-intelligence): AI-powered BI that lets people ask questions about company data in plain language instead of building a dashboard first. - [Company Brain — definition | Libra AI](https://trylibra.ai/glossary/company-brain): The connected layer of company knowledge, context and systems that gives AI the information it needs to reason and act across the business. - [Connector — definition | Libra AI](https://trylibra.ai/glossary/connector): The link between an AI platform and a system like Slack, Gmail or Salesforce. - [Context Graph — definition | Libra AI](https://trylibra.ai/glossary/context-graph): A structured map of how a company’s information connects: documents, decisions, people, conversations. - [Context Layer — definition | Libra AI](https://trylibra.ai/glossary/context-layer): The part of an AI system that holds a company’s connected knowledge for every agent to reason from. - [Context Window — definition | Libra AI](https://trylibra.ai/glossary/context-window): The amount of text an AI model can process in a single request, and why it is not the same as persistent memory. - [Context-Aware Reasoning — definition | Libra AI](https://trylibra.ai/glossary/context-aware-reasoning): An AI system’s ability to interpret a request using surrounding company knowledge, not just the prompt. - [Data Residency — definition | Libra AI](https://trylibra.ai/glossary/data-residency): The requirement that company data is stored and processed in a specific country or region, and not moved outside it. - [Desktop AI — definition | Libra AI](https://trylibra.ai/glossary/desktop-ai): AI embedded directly in an employee’s everyday computer workflow, rather than a separate app to visit. - [Diarization — definition | Libra AI](https://trylibra.ai/glossary/diarization): Working out who spoke when in a recording, so a transcript is attributed rather than one undifferentiated block of text. - [Embedding — definition | Libra AI](https://trylibra.ai/glossary/embedding): A numeric representation of what content means, letting a system compare passages by sense rather than by the words they use. - [Enterprise AI — definition | Libra AI](https://trylibra.ai/glossary/enterprise-ai): AI systems built for the access controls, deployment and scale of a company, not a consumer tool. - [Evaluation (Evals) — definition | Libra AI](https://trylibra.ai/glossary/evaluation): Measuring an AI system’s output against known-good cases, so a change can be shown to help rather than assumed to. - [Function Calling — definition | Libra AI](https://trylibra.ai/glossary/function-calling): How an AI model invokes an external tool or action as part of generating a response. - [Governed Execution — definition | Libra AI](https://trylibra.ai/glossary/governed-execution): An AI agent taking real action across systems, but only within defined permissions and approvals. - [Human-in-the-Loop — definition | Libra AI](https://trylibra.ai/glossary/human-in-the-loop): A workflow where a person reviews or approves an AI system’s output before it takes effect. - [Identity Graph — definition | Libra AI](https://trylibra.ai/glossary/identity-graph): How a person’s activity, permissions and role connect across a company’s different systems. - [Knowledge Base — definition | Libra AI](https://trylibra.ai/glossary/knowledge-base): The indexed store of a company’s documents and decisions that an AI system draws on. - [LLM (Large Language Model) — definition | Libra AI](https://trylibra.ai/glossary/llm): The underlying AI system trained on text to generate and reason about language: the model, not the platform built around it. - [MCP (Model Context Protocol) — definition | Libra AI](https://trylibra.ai/glossary/mcp): An open standard that lets AI systems connect to external tools and data sources in a consistent way. - [Meeting Bot — definition | Libra AI](https://trylibra.ai/glossary/meeting-bot): The participant an AI platform sends into a call, so the recording and transcript do not depend on someone remembering to press record. - [Multi-Agent System — definition | Libra AI](https://trylibra.ai/glossary/multi-agent-system): Several narrow agents coordinating on one outcome, instead of a single agent holding every tool and every instruction. - [Natural Language Query — definition | Libra AI](https://trylibra.ai/glossary/natural-language-query): Asking a question of company data in ordinary words and having the system work out the query, instead of writing one. - [Orchestration — definition | Libra AI](https://trylibra.ai/glossary/orchestration): The coordination of multiple AI agents or steps toward a single outcome, rather than one agent handling one task alone. - [Permissions-Aware AI — definition | Libra AI](https://trylibra.ai/glossary/permissions-aware-ai): A system that enforces the same access rules as the tools it connects to. - [Persistent Company Memory — definition | Libra AI](https://trylibra.ai/glossary/persistent-company-memory): An AI system’s ability to retain and reuse context across sessions, not start from zero. - [Query Understanding — definition | Libra AI](https://trylibra.ai/glossary/query-understanding): Working out what a question actually asks — the entity, the timeframe, the intent — before anything tries to retrieve an answer. - [RAG (Retrieval-Augmented Generation) — definition | Libra AI](https://trylibra.ai/glossary/rag): A technique where a model retrieves relevant information before generating a response. - [Semantic Search — definition | Libra AI](https://trylibra.ai/glossary/semantic-search): Search that matches meaning rather than wording, so a result can be relevant without sharing any words with the question. - [Structured Output — definition | Libra AI](https://trylibra.ai/glossary/structured-output): Constraining a model to return data in a fixed shape, so the result can be used by a program rather than read by a person. - [System of Record — definition | Libra AI](https://trylibra.ai/glossary/system-of-record): The authoritative source for a specific type of company data, like a CRM for customers. - [Task — definition | Libra AI](https://trylibra.ai/glossary/task): A unit of work delegated to an AI system and tracked to completion, rather than a question answered and forgotten. - [Tenant Isolation — definition | Libra AI](https://trylibra.ai/glossary/tenant-isolation): The guarantee that one customer’s data and compute cannot reach another’s inside a shared platform. - [Token — definition | Libra AI](https://trylibra.ai/glossary/token): The unit a language model reads and writes — roughly a word fragment — and the unit AI usage is measured and billed in. - [Trigger — definition | Libra AI](https://trylibra.ai/glossary/trigger): The event that starts a piece of automated work — a message arriving, a record changing, a meeting ending — instead of a person starting it. - [Unstructured Data — definition | Libra AI](https://trylibra.ai/glossary/unstructured-data): The documents, emails, messages and recordings that hold most of what a company knows and fit no schema. - [Vector Database — definition | Libra AI](https://trylibra.ai/glossary/vector-database): A store that indexes content by meaning rather than keywords, so an AI system can retrieve passages that are relevant without matching the wording. - [Work AI — definition | Libra AI](https://trylibra.ai/glossary/work-ai): AI systems built to complete tasks across a company’s tools, not just answer questions. - [WorkBase — definition | Libra AI](https://trylibra.ai/glossary/workbase): Libra’s layer for turning company context into workflows that run across your connected systems, with a person signing off where it matters. - [XAI (Explainable AI) — definition | Libra AI](https://trylibra.ai/glossary/explainable-ai): Making an AI system’s output inspectable — what it drew on and what it did — rather than asking people to trust it. - [Zero Data Retention — definition | Libra AI](https://trylibra.ai/glossary/zero-data-retention): An arrangement where a provider processes a request and stores nothing from it once the response is returned. ## Legal - [Terms of Service | Libra AI](https://trylibra.ai/terms): Read Libra AI's Terms of Service covering the terms and conditions that apply when using Libra products and services. - [Privacy Policy | Libra AI](https://trylibra.ai/privacy): Read Libra AI's Privacy Policy to understand how information is collected, used, protected, and handled across Libra services. - [Cookie Policy | Libra AI](https://trylibra.ai/cookie-policy): Read Libra AI's Cookie Policy to understand how cookies and similar technologies are used across the Libra website and services. - [Data Processing Agreement | Libra AI](https://trylibra.ai/dpa): Review Libra AI's Data Processing Addendum covering data processing terms and responsibilities for customers using Libra services. - [Fair Use Policy | Libra AI](https://trylibra.ai/fair-use): Read Libra AI's Fair Use Policy covering responsible and acceptable use of Libra products, services, and platform resources.