Quickstart
Connect your coding agent, scaffold a loop, and run it locally before it costs anything.
Open the quickstartEverything you need to build agents that ship.
Documentation, templates, guides, and every post we have published, in one index.
where to go
Six places to start.
Docs
The canonical reference for every tool, capability, and change Sapiom ships.
Read the docsGuides
How to author a step graph, shape shared state, and control what a run costs.
Read the authoring guideTemplates
Three agent loops with every step already wired, each cloned into a repository you own.
Browse the templatesBlog
Practical guides, product thinking, company news, and conversations with the people building Sapiom.
Read the blogCommunity
The SDK, the MCP server, and the example agents are all public.
View the org on GitHubStart from a template. Own the code.
Open a proven workflow, adapt it to the job, and keep every generated TypeScript step inspectable.
Review entitlements against role and employment state, then route removals or exceptions for approval.
Resolve a case from authoritative context or hand it to a person with the full decision trail.
Create a cited account brief, relationship graph, and review-ready next actions from live and internal sources.
in the client you already use
Cloned, checked, and run locally without capability charges.
Your coding agent clones the loop into a repository you own, validates the graph offline, and runs the real step code with Sapiom capability calls served by local stubs.
> Clone the web research digest template and run it locally ⏺ sapiom_dev_agents_clone(web-research-digest) ⍻ ordinary TypeScript · search → summarize ⏺ Check(sapiom.json) · offline ⍻ graph valid · 2 steps · entry search · terminal summarize ⏺ MCP(sapiom_dev_agents_run_local) ⍻ Sapiom capability calls stubbed · no capability charge ⏺ Local run passed. The search result was stubbed and the digest returned. ✓ ready to deploy · sapiom_dev_agents_deploy when you are > ▁
> Run Web Research Digest locally ⏺ sapiom_dev_agents_clone(web-research-digest) ⏺ Check(sapiom.json) · graph valid ⏺ MCP(sapiom_dev_agents_run_local) ✓ Local run passed no capability charge > ▁
from the blog
Latest writing.
Field notes, launch stories, and practical guides from the team.
The Next Trillion Agents
Announcing Sapiom’s $35 million Series A and the production infrastructure we spent eleven months building.
Read the story
Sapiom Raises $35 Million Series A to Power the Next Trillion AI Agents
Sapiom removes the infrastructure barriers to shipping and scaling agents, starting with the cost of execution.
Read articleLearn by shipping.
The fastest way to understand Sapiom is to run an agent locally, then deploy when you are ready.