Everything you need to build, ship and govern agentic systems.
Documentation, quickstarts, deep guides, whitepapers, templates and a full security & trust center — written by the forward-deployed teams who design and run agentic systems in real businesses, every day.
The agentic business system: an architecture for governed autonomy
How a team of governed AI agents can structure operations, ship software and run the daily work of a business — without giving up control. We break down the six platform primitives, the governance wedge, and the operating model that has been proven across 180+ businesses.
Inside: the Mainspring runtime, the Loom fabric, Cadence orchestration, and how Aegis, Warrant and Ledger answer the four questions that decide whether autonomy is safe — who authorized it, what it touched, can you prove it, and can you stop it.
What you'll take away
A reference architecture
The full diagram of how runtime, fabric, orchestration, governance, identity and the system of record fit together.
A governance model that scales
Authorization, scoping, provable audit and a hard stop — designed in from the first agent, not bolted on later.
An adoption playbook
How forward-deployed teams move a business from a single agent to a governed fleet, with the metrics to prove it.
The full library.
Nine ways into the platform — from your first ten minutes to a board-ready trust review.
Documentation
The complete reference for the platform and its six primitives — concepts, configuration and the mental model behind governed autonomy.
Open docs →Quickstarts
Stand up your first governed agent in under ten minutes. Copy-paste recipes for the most common first builds.
Get started →Guides
Step-by-step walkthroughs for structuring operations, shipping software and putting agents to work on real processes.
Read guides →Whitepapers
Deep, citable papers on the architecture, the governance wedge and the economics of agentic operations.
Browse papers →Webinars
Live and on-demand sessions with our forward-deployed teams — architecture deep-dives, customer stories and office hours.
Watch sessions →Templates
Production-ready agent blueprints, orchestration patterns and governance policies you can fork and deploy.
Get templates →API reference
Every endpoint, schema and SDK call for Mainspring, Cadence and the Ledger — with authenticated examples.
View reference →Changelog
Every release, primitive update and breaking change — versioned, dated and written for engineers.
See what's new →Security & Trust center
SOC 2 Type II, GDPR, sub-processors, data residency and our complete answer to the four governance questions.
Open trust center →Start with the ones teams reach for first.
Hand-picked, end-to-end walkthroughs that take you from idea to a governed agent running in production.
Ship your first governed agent
From a blank workspace to a running agent with scoped permissions, an audit trail in the Ledger and a working stop switch.
Read guide →Design an orchestration with Cadence
Wire multiple agents into a reliable, observable workflow — retries, handoffs and human-in-the-loop checkpoints.
Read guide →Set up Warrant identity & access
Give every agent a scoped identity, least-privilege access and a verifiable answer to "who authorized this?"
Read guide →Make audit provable with the Ledger
Turn every agent action into a queryable, tamper-evident record — and produce a board-ready audit in minutes.
Read guide →Migrate a workflow off n8n
Move an existing automation onto a governed agentic system — with the audit, identity and control you were missing.
Read guide →Automate routine work, safely
Identify the 70% of routine tasks worth handing to agents, and the guardrails that keep that handover trustworthy.
Read guide →Reference docs for every primitive.
Each part of the platform has its own home in the documentation — concepts, configuration and copy-paste examples.
Mainspring ™
Where agents run. Lifecycle, isolation and execution docs.
Loom ™
How agents, tools and data connect. Integration patterns.
Cadence ™
Multi-agent workflows, handoffs and human checkpoints.
Aegis ™
Policy, scoping and the hard stop. The control surface.
Warrant ™
Scoped agent identity, access and authorization.
Ledger ™
Provable, queryable audit of everything every agent did.
Proof before trust.
Autonomy is only safe when you can answer four questions about every action. Our trust center documents how lastloop answers all four — with certifications, sub-processor lists and data-residency controls you can hand straight to your security team.
Who authorized it
Every action traces to a scoped identity and an explicit grant in Warrant.
What it touched
Full visibility into the data, tools and systems each agent reached.
Can you prove it
Tamper-evident records in the Ledger, ready for audit and review.
Can you stop it
A real, immediate hard stop across the whole fleet via Aegis.
What shipped recently.
Conditional handoffs in workflows
Route work between agents based on Ledger state, with full audit of every branch taken.
Read release →Fleet-wide stop switch, <1s
Halt every running agent across an organization in under a second, with a recorded reason.
Read release →Exportable audit bundles
One-click, signed audit exports scoped to a date range, agent or workflow.
Read release →Read the docs, or let our teams build it with you.
The platform is yours to build on. When you want it done faster, the forward-deployed teams of lastloop GmbH · Business Solutions design, ship and operate your agents alongside you — proven across 180+ businesses.