Empresa como código.Um framework de backend e um servidor para agentes.
Um espaço de trabalho é sustentado por um repositório git. Cada agente é um agente pi personalizado: um harness mínimo e open source que você estende com extensões do pi. kortix.yaml declara os agentes, conectores, gatilhos e regras. O Kortix os implanta na nuvem e executa cada sessão em seu próprio computador. Crie sua própria experiência com @kortix/sdk e @kortix/ui.
One block per agent. It gets the tools, connectors and skills the block lists, and nothing else.
- # Engineering
- eng-triage:
- description: Triages every new GitHub issue.
- prompt_file: agents/engineering/triage.md
- tools: [read, grep, glob, web_search]
- connectors: [github, kortix_slack]
- skills: [triage-issue]
- extensions: [triage-policy]
eng-triage
Faz a triagem de cada nova issue do GitHub.
- read
- grep
- glob
- web_search
- github
- kortix_slack
- triage-issue
- triage-policy
O repositório em /workspace, em seu próprio branch
Um arquivo em um repositório. Dosco runs what it declares.
Quatro ideias. The whole framework, in four parts.
A project is a git repoYour company, versioned.
- Dosco hosts it, or link GitHub
- Agents, skills and memory are files
- Every change is a commit
kortix.yaml declares itOne file, checked before it runs.
- Agents, connectors, triggers, rules
- Access is deny-by-default
- kortix validate on ship and merge
Custom pi agentsA minimal harness, deployed to the cloud.
- pi, a modular agent runtime
- pi extensions from npm or the repo
- A computer and a branch per session
Your own experienceThe same backend, from code.
- @kortix/sdk: one typed client
- @kortix/ui: the web app’s components
- A CLI and a REST API
Como funciona.One part at a time.
Agentes
An agent is a block in kortix.yaml and a Markdown prompt. The block names its model, tools, connectors and skills.
Access is deny-by-default. An agent gets no connector, secret or skill the block does not list.
kortix.yaml
# Marketing marketing: description: Drafts LinkedIn posts from shipped work. prompt_file: agents/marketing/content.md tools: [read, web_search] connectors: [linkedin, kortix_slack] skills: [write-post] commands: [draft-post] extensions: [web-access]agents/marketing/content.md
You write Arcadia LinkedIn posts from work the team shipped.Follow the write-post skill and the tone in the company memory.Every post waits for a person to approve it.A arquitetura. One API. A computer per session.
Clients
People and code reach Dosco the same way.
- Web app
- Desktop
- Mobile
- CLI
- Slack
- Microsoft Teams
- Your app
Dosco API
Decides who may run which agent, starts sessions, holds connector logins and change requests.
Workspace repository
kortix.yaml, agents, skills and memory. Dosco hosts it, or you link GitHub.
Computador de sessão
One per session. Your pi agent and its extensions work here, on the session’s own branch.
LLM gateway
Every model call. Managed models or your own keys.
Drive
Shared folders and company memory, open at /drives.
Dúvidas sobre o framework.Answered from the docs.
Anything else, write to the team.
Do I have to write YAML?
No. People manage agents, connectors and triggers under Customize in the web app, and Create in chat writes a new agent with you. Those edits land in the same kortix.yaml.
Can I keep the repository on GitHub?
Yes. With a GitHub origin, kortix ship links that repository, and plain git push and git pull work. Without one, Dosco creates and hosts the repository.
What does kortix validate check?
Every key against the schema, then the files around it: prompt files that do not exist, skill folders without a SKILL.md, and connector or secret names the file does not declare. It exits 1 on an error. kortix ship runs the same checks, and a merge with an invalid kortix.yaml is refused.
When does a change to kortix.yaml take effect?
When it lands on the base branch, through a merge or a push. New sessions start with it. A running session picks it up at the start of its next turn.
What runs inside a session?
pi, a modular agent runtime. Every agent you declare is a pi agent, and every new session runs it on its own cloud computer. It calls models only through the Dosco LLM gateway.
Can I extend the agent harness?
Yes. Write a pi extension in TypeScript, or pin a pi package from npm, and register it under extensions in kortix.yaml. Each agent loads only the extensions it lists. An extension can add tools, such as your own policy as code, and react to events, such as a check before every tool call.
Where do secret values live?
In Dosco, encrypted. kortix.yaml lists only the names a project needs. No page, command or API shows a stored value again.
Can I run it myself?
Run Dosco on Cloud or request a dedicated single-tenant VPC deployment for enterprise isolation.

