Agents, skills, tools, connectors and schedules are files in a repo. Edit them in your IDE, ship them with one command.
$ curl -fsSL https://kortix.com/install | bashkortix_version: 2
default_agent: support-triage
project:
name: acme-ops
agents:
support-triage:
file: agents/support-triage.md
connectors: [slack, linear]
connectors:
- slug: slack
provider: composio
app: slack
- slug: linear
provider: composio
app: linear
triggers:
- slug: daily-triage
type: cron
agent: support-triage
cron: "0 0 8 * * *"
prompt: "Triage new support tickets and flag the urgent ones."Every agent runs in its own disposable cloud VM, on its own git branch. Spin up thousands in parallel. Nothing is shared between runs.
Every agent runs in its own disposable cloud VM, on its own git branch. Spin up thousands in parallel. Nothing is shared between runs.
Agents, skills, triggers, connectors and policies are plain files in one repo. Diff them, review them in a change request, roll them back.
---- model: anthropic/claude-opus-4-8+ model: openai/gpt-5 tools: lookup_order: true
Open source. Self-host the exact same stack, bring your own runtime and model keys. No black box, no lock-in.
Agents live in your repo as files. Kortix runs each session in its own sandbox.
kortix init adds kortix.yaml, agents/ and skills/ to your repo.
A persona, a model and its tools in one file you can review.
kortix ship pushes, builds the sandbox and asks for missing secrets.
Every session runs in its own cloud VM, on its own branch.
Agents work on branches. You review and merge.
Pick the model per agent, or bring your own coding agent.
Every session runs in its own isolated sandbox on its own branch off main. Run them in parallel; nothing is shared.
The same binary is pre-authenticated inside every sandbox, so agents use it too.
3,000+ apps, plus any MCP, OpenAPI, GraphQL or raw HTTP endpoint, behind one Connector interface your agents call.
KORTIX_TOKENconnectors:kortix connectors