Daily briefing
Every morning before the team logs on, pulls together the day's calendar, the overnight emails worth flagging, and the open tasks each person owns. Posts a one-screen brief into Slack.
Posts to: Slack
The AI agentworkspacefor any team
You write the rules in plain language and pick what the agent can connect to. It runs the job in an isolated workspace and brings the result back where your team already works. Every step is on the record.
Be the first in lineNo spam. unsubscribe anytime
We host itstart in minutesYou host itdata stays put
01 · How it works
Turn recurring work into an operating policy. Then let agents run it
d5s turns a business process into a governed agent: rules, permissions, tools, memory, approvals, and audit trail. Compliance can review it; team leads approve it, before any action runs.
Rules
Plain-language rules
You write what the agent is for, what it can read, what it can write to, and when it should stop and ask a human. Compliance reads it. Team leads approve it.
Connections
Plugs into your tools
Connect a database, a SharePoint folder, a Slack channel, an internal tool. Each connection is read-only, write, or off. The agent only touches what you allow.
Workspace
A clean workspace for every run
Every time the agent runs, it gets its own isolated environment. Nothing leaks across conversations, teams, or customers. The workspace is destroyed when the work is done.
Evolves with use
Memory and playbooks
Memory carries facts across runs. Once a job runs cleanly, the steps are saved as a playbook. The agent watches cost, accuracy, and feedback, and proposes its own improvements.
02 · Runtime
Automations that improve over time
Agents start as reasoning loops and earn their way into fast, auditable playbooks. When something breaks, they patch themselves and propose a solution back to you.
First runs
Reasoning loop
The agent reads the rules, picks its tools, and works through the problem from scratch. Slower, and pays the full reasoning cost while it finds a clean path.
→
After clean runs
Playbook
The working path becomes a fixed playbook. The same job runs in seconds at a fraction of the cost, with steps that stay easy to review.
→
When something changes
Recursive refinement
If a column gets renamed or an API moves, the agent detects the drift, patches the playbook, and asks a human to approve the change.
03 · Security
Security-first. Compliance-ready
Built for regulated workloads. Two ways to deploy: managed by us for speed, or in your own cloud when data can’t move. Every pillar below applies to both.
01
Isolated workspace
Every conversation gets its own clean, sealed environment. It is destroyed when the work is done. Nothing leaks between runs, customers, or teams.
02
BYOC
Bring your own cloud. The platform runs in your account, in any region. Your data stays next to the systems agents need to reach.
03
SSO
Plug into the identity provider your team already uses. Set who can run what. Every read, every write, every decision is logged for audit.
04
Self-hosted
Same product whether we host it or you host it. Bring your own login system and your own AI models, including air-gapped setups where no traffic leaves your network.
FAQ
The short version
What is d5s?+
Use d5s to set up AI agents that handle repetitive tasks for your whole team, like clearing the support queue each morning, tracking suppliers, or pulling together the weekly report. Each agent plugs into your existing tools, follows rules you write in plain language, and logs every step.
Who uses d5s?+
Teams that want AI doing real work, not just answering questions. Best fit: legal, finance, life sciences, infrastructure, and government suppliers, where data sensitivity rules out hosted chatbots. Also a fit: any data-heavy team that wants one tool that runs hosted or inside their own cloud.
How do I try d5s?+
Join the waitlist. We onboard new teams with a short call, help define your first workflow, and connect it to one of your systems in the first session.
How is this different from a hosted chatbot?+
Hosted chatbots stop at the reply. With d5s, agents run in isolated sandboxes with scoped connectors to your real systems: reads logged, writes gated, every action under a policy your compliance team signed off on.
Where does my data live?+
Depends on who hosts it. When we host it: in AWS, in the region you pick, isolated per customer, with infrastructure, identity, and model access handled for you. When you host it: inside your own cloud, in any region, behind your own login system. Data never leaves the environment you deploy into.
Which models do you support?+
You pick. We handle access to OpenAI, Anthropic, and others, or you bring your own keys. Each agent is locked to a specific model, so switching providers is always explicit and reviewable.
What does self-hosting actually require?+
One package covers the control plane, sandboxes, identity, and connectors. Deploy next to the systems your agents need to reach.
Can I bring my own identity provider and models?+
Yes. Self-hosted setups work with your existing identity provider and your own models, including air-gapped deployments where no traffic leaves your network.
What can agents actually touch?+
Only the things you connect, with the permissions you set. Databases, SharePoint folders, Slack channels, internal tools. Each connection is read-only, write, or off, and you can change or revoke it any time.
Examples
What teams hand off to an agent
Automate any repetitive task. Each agent connects to your existing tools, follows rules you write in plain language, and posts results where your team already works.
Latest changes
What we've shipped to the workspace recently.
Sub-agents go to work — two profiles, real sessions, and connector access
Sub-agents now have explorer and generalist profiles, appear as real sessions with streaming and usage tracking, and can use connectors to gather context.
A public changelog — and it writes itself
The d5s changelog is live at /changelog, updated daily by a d5s automation that reads what shipped and drafts public release notes.
Chat gets fast — token streaming and smarter scrolling
Chat now streams tokens incrementally for sub-second first-word latency, pins new messages to the top, and scrolls smoothly so you never lose your place.
Tell us the work you’d hand off
We’re working with our first teams now. Tell us the job you’d want an agent to do, and we’ll show you what it would look like.