GDPR
PrivacyControls for your data
Knowledgebase + AI Workspace
Every source. One context. Every AI.
Connect knowledge once. Put 24/7 agents to work. Protect sensitive data before it reaches the model, with central rules for roles, skills and plugins.
EU AI ACT
Responsible AIAuditable AI actions
ISO 27001
InfrastructureWithin the agreed hosting scope
EU
Data residencyEuropean residency options
Two products. One knowledge advantage.
Better knowledge. More capable AI.
Start with the Knowledgebase. Put your knowledge to work in your existing AI tools or add the integrated AI Workspace.
The standalone foundation
Knowledgebase
Connect your sources. Keep knowledge current. Find reliable context, with an integrated wiki and the right access controls.
- Enterprise search with source evidence
- Connected sources and an internal wiki
- For people, APIs and MCP
The integrated extension
AI Workspace
Turn context into results with model choice, agents and automations in one centrally managed workspace.
- Chats and agents with company context
- MCPs, skills and recurring tasks
- Central model and action policies
More than an AI chat
Around the clock. Under your control.
24/7 agents, privacy mode and centrally managed capabilities work together. You decide which knowledge, tools and actions are allowed.
24/7 agents. Work beyond the open tab.
Prepare a morning briefing, monitor a market or follow up on new documents. Agents and automations run on the server, on a schedule or configured event, even when you are not in the workspace.
Use schedules, email, webhooks, HTTP polling, RSS and knowledge or connector events. Permissions, approvals, runtime limits and available model budgets still apply.
Privacy mode. Less sensitive data for the model.
The AI privacy layer can replace detected sensitive text with placeholders before model transmission. Administrators can make protection optional or mandatory. Knowledgebase policies also protect supported AI retrieval and MCP paths.
Your authorized human view stays readable. Detection is not a guarantee of anonymity. Approved external tools may receive original values; their own data-processing terms still apply.
Skills and plugins. Centrally governed.
Provide a shared library of skills and MCP connections. Decide which roles may use each resource and whether employees can add personal connections, instead of configuring every employee separately.
Source permissions and action approvals remain separate checks. A plugin release does not grant access to every document or override a blocked action.
A browser that works with your agent.
Research websites, follow pages and work with web interfaces in an isolated agent browser. Inspect its steps alongside the conversation instead of switching between disconnected research tools.
Browser access follows configured policies. Isolation is not anonymity or guaranteed access: website restrictions, logins and required approvals still apply.
One chat, from answer to action.
Ask a short question or describe a multi-step task without choosing a separate agent mode. The assistant can answer directly or use permitted tools when the task calls for them.
An isolated execution environment starts only when needed. Dedicated agents keep their own settings and workspace; chat routing does not replace those controls.
Memory at the right level.
Current task context, an agent’s saved workspace and instructions, optional personal memories across chats, and permission-aware Knowledgebase access serve different purposes. Keep useful context without treating everything as shared knowledge.
Control personal memories in settings and optionally store them as editable Markdown in your own private workspace. Scheduled jobs use their agent’s configured instructions and resources.
The complete stack
Connect once. Put context to work.
Your sources stay connected. Your knowledge becomes usable. People and agents work from the same foundation.
02AI WorkspaceOptional extension
Chat
Agents
Automations
Permission-aware context
01KnowledgebaseWorks independently
Search & graphIntegrated wiki
Continuous sync
Context makes the difference
Your AI should not start from scratch. Your company does not.
Current documents, established decisions and internal expertise give AI work its foundation. CogLake makes that knowledge accessible, discoverable and useful to the right people and agents.
Explore the shared foundationAI Workspace
Beyond answers. Into action.
From the first conversation to a recurring task: shared context, clear permissions and results you can inspect.
Research you can work with.
An agent researches, checks sources and prepares a grounded draft. You decide what happens next.
A briefing before the day begins.
A personal agent works with its workspace and prepares recurring updates on the topics that matter to you.
Routine, with the right checkpoint.
A trigger starts the task. Tools carry out permitted steps. Approvals keep you involved in consequential actions.
From knowledge to workExample workflow
- Brief
Summarize the changes to our product offering and prepare a customer briefing.
- Context
Read authorized product documents and current notes from the wiki.
- Result
Create a sourced briefing and make it available for review.
Control throughout the workflow
More capability. Clear boundaries.
A consistent governance model for your knowledge and AI work, from the first lookup to the action that follows.
Knowledge with access controls
Source and workspace permissions determine which information an employee or agent may retrieve.
Central model and tool policies
Approve providers, MCPs and skills. Decide which personal extensions are allowed.
Deliberate action approvals
Approval modes, allow/deny rules and audit make agent work inspectable.
Model freedom
Change the model. Keep the knowledge.
CogLake connects its model offering, your own API keys and self-hosted models to the same knowledge foundation. Your organization decides which models employees may use.
Models and providersModel providers
ChatGPT
Claude
Gemini
Mistral AI
ChatGPTClaudeGeminiMistral AIGoverned contextOne policy boundary
Interfaces
Search
AI agents
MCP & API
Local models
Company context for your teams’ AI tools – through API and MCP
ChatGPTClaudeGemini
GitHub CopilotMistral AIBenchmark-tested retrieval quality
Find the knowledge. Measure the quality.
Public datasets and recorded tests make retrieval quality, source permissions and index consistency inspectable.
Good to know
The questions to ask before you start.
Do I need the AI Workspace?
No. The Knowledgebase can be purchased and used independently, including its search, wiki and authorized API/MCP access. AI Workspace is an optional integrated extension.
How do chat, agents and automations differ?
In chat, the assistant can answer directly or use permitted tools for multi-step tasks, without a separate mode switch. A personal agent retains its workspace and memory for ongoing work. Automations start defined tasks on a schedule or event.
Can employees use their own providers?
If company policy allows it, employees can add personal API keys and resources. Administrators can restrict provider access and offer centrally managed connections instead.
Does an agent act without asking?
That depends on the selected approval mode, allowed and denied actions, and organization policies. Required approvals appear in the conversation; central restrictions cannot be bypassed by the user mode.
Is model usage included in Knowledgebase pricing?
Knowledgebase works independently. Add AI Workspace for chats, agents and automations, with monthly model usage included per user. Additional usage requires authorised credit. BYOK usage is billed by your connected provider.
Your next step
Start with knowledge. Go further with AI.
Knowledgebase on its own or with AI Workspace: find the right starting point for your organization.
Book a walkthrough