Agents
Specialized agents with their own remit, their own tools, and their own memory.
The platform Pre-configured AI agents.
Pre-configured AI agents.
Productive from day one.
Every agent has a clear remit, its own workspace, tested skills, and the right model. Centrally configured by IT and the business, rolled out under control.
A complete AI agent stack. On one sovereign, Swiss-hosted platform.
AIgent is the sovereign AI agent platform from Switzerland. Agents, skills, knowledge, model choice, and audit trail work as one, centrally configured by IT and the business, rolled out under control, hosted in Switzerland or on-premise.
ClaudeGlass box
Every step traceable and auditable. Usage and cost per agent, without surveillance.
Skills
Tested capabilities over real code instead of estimated text.
Knowledge
Grounded answers from your own documents, with sources.
Model choice per agent
The model follows the sensitivity of the data.
Routines
Agents that run on a schedule, via webhook, or API on their own.
Roles & governance
Organizations, groups, roles, and API keys, cleanly separated.
Sovereignty
Swiss hosting or on-premise. Your data stays yours.
Integrations
Connected to the tools you already work in. Over 50 integrations, ready to use.
A tour through the platform.
From the specialized agent to the complete audit trail. This is what the work AIgent does for you looks like.
Agents
Your AI workforce, specialized per task.
Every agent has a clear remit, its own tools, and its own memory. Like a team member who is set up on day one and picks up right where you left off.
- Its own remit, system prompt, and guardrails
- Its own tools, skills, and one model per agent
- Memory across conversations, even days later
Chat & model choice
Every model. One interface.
The model follows the sensitivity of the data. International models for non-critical tasks, Swiss-hosted or on-premise models for sensitive data. Set per agent.
- Claude, GPT, Gemini in one interface
- Swiss-hosted and on-premise models
- Answers from your company knowledge, with sources
Skills
Reliable capabilities. No guessing.
Skills are tested code, not just text. Same input, same output. Install from the catalog, bring your Claude Code or Codex skills, or write your own.
- Tested code instead of estimated text
- Claude Code, Codex, and any open format
- Install from the catalog or build your own
Knowledge
Answers from your data, with sources.
Upload documents, spreadsheets, and websites. AIgent organizes everything in seconds. Your agents search it before they answer, cited to the passage that is actually yours.
- PDF, Word, Excel via drag and drop
- Lightning-fast search with source citations
- One knowledge base, shared across many agents
Transparency & oversight
No black box. A glass box.
One dashboard across every agent: what they deliver, which model they use, how long they take, and what they cost. Steered by agents and tasks, not by people.
- Plan, model, tool, and result visible
- Complete audit trail for oversight and audit
- KPIs without surveillance, privacy-compliant
Centrally configured, rolled out under control.
From use-case to a productive agent in three steps. The business uses AI, control stays central with IT and compliance.
1
Tailored to your infrastructure
On-premise hosting, Entra ID and other auth providers, Swiss-hosted models or bring-your-own-model.
On-premiseEntra IDSwiss ModelsBYO Model
2
Tailored to your users
Role and permission synchronization via Entra ID or your existing directory services. Access follows your organization.
RolesPermissionsSync
3
Tailored to your needs
Pre-prompted agents for concrete tasks. The right model, tool set, and security level are chosen per agent.
PromptsToolsModel choice
Common platform questions.
- What is an AI agent platform?
- An AI agent platform provides autonomous agents that do real work instead of just chatting. AIgent gives every agent its own workspace, tested skills, access to your company knowledge, and a model per task, run in Switzerland or on-premise.
- Does AIgent do real work or just chat?
- Real work. Agents write emails, produce reports, review documents, and reconcile data through tested code (skills) instead of just text. Every step is traceable in the audit trail.
- Where is my data processed?
- In Swiss data centers or entirely on your own infrastructure, up to air-gapped. No US Cloud Act exposure, off US hyperscalers. revFADP and GDPR compliant out of the box.
- Which AI models can I use?
- You choose the model per agent: Swiss-hosted or on-premise models for sensitive data, or international models such as Claude, GPT, and Gemini for non-critical tasks. You can connect your own models too.
- Do I need developers to set up agents?
- No. Agents are built without coding. IT and the business configure model, tools, and rights centrally, and the business starts with a click. The first agent is live in under 15 minutes.
- Is the platform auditable and compliance-ready?
- Yes. The glass box records every step: plan, model, tool call, tokens, cost, and result. KPIs are reported without surveillance. Designed for revFADP, GDPR, EU AI Act, and DORA.
Sovereign AI. For every discerning company.
On your use-case, in your environment, within a clearly defined scope. See measurable results within a few weeks.
Secure early access. No per-seat fees. No lock-in.