AI orchestration & operations
Coordinate autonomous AI work
like a control plane should.
Orchicon orchestrates. Runtimes execute. One binary, one command, and your local stack — Postgres, NATS, the Grafana telemetry plane, the SPA — is up. Or run the whole thing in a single container.
$ curl -fsSL https://orchicon.dev/install | bash
PS> irm https://orchicon.dev/install.ps1 | iex
One command, full setup: downloads the orchicon binary,
pulls the published images, starts the runtime daemon, launches the
single-container instance, and prints how to connect / start / stop.
Re-running updates to the latest release. On Windows the whole stack
runs inside WSL2 — the installer provisions it and
installs the Linux binary into the distro. See
options below.
Quick start
-
1
Install
Run the one-liner above. Requires
curlandtar(or PowerShell on Windows — the Windows path runs the stack inside WSL2). -
2
Start the stack
$ scripts/container.sh up devBoots Postgres, NATS, Grafana (Tempo + Loki + VictoriaMetrics), the OTel collector, runs migrations, and serves the control plane + embedded UI on localhost:8080. Requires Docker.
-
3
Log in & build
Open the URL, sign in with the built-in dev IdP, create a project, define a Worker, and dispatch a Task. The whole end-to-end flow is verifiable locally.
What Orchicon is
An orchestration platform for autonomous AI work. It coordinates the moving parts — projects, workers, scheduling, policies, telemetry, recovery, governance — and talks to pluggable runtimes (OpenCode first) over gRPC.
Projects & work
Tenants, projects, work items as a DAG (Epic → Feature → Task → Subtask), with cycle-detected dependencies.
Workers & dispatch
Versioned Worker definitions (system prompt, permissions, gated tools, budgets). Reconciler-driven dispatch to healthy adapters.
Workflows
Step-DAG workflows: task, decision, approval, parallel, recover. Visual React Flow editor. Real-time step transitions.
Recovery engine
Default 6-step recovery workflow (capture → summarize → preserve → review → plan → resume) with checkpoint replay and bounded auto-relax.
Rego policy
OPA v1 at the dispatch decision point. Narrowest-scope-first evaluation, full Rego trace per decision for explainability.
Telemetry & cost
OTel → Grafana stack (Tempo + Loki + VictoriaMetrics). Per-tenant cost roll-up (Tenant → Project → Task → Execution) with streamed usage events.
Single container + per-workflow runtime containers
The whole stack — Postgres, NATS, telemetry, control plane — runs in one container with orchicon container (PID-1 supervisor). Dev + prod as two containers, your data preserved.
Worker executions run in their own short-lived container per workflow run (orchicon runtime-daemon on the host), armed with a baked toolchain and running as your user — no root, fully ephemeral, torn down at completion.
Orchicon orchestrates. Runtimes execute.
The control plane is a single Go binary. The runtime is a
sidecar. Adapters never touch Postgres or NATS directly — they
stream telemetry over gRPC. Swap OpenCode for something else
by implementing orchicon.adapter.v1.
Build from source
Pre-built binaries ship from GitHub Releases. Until the first release is cut, build from source:
$ git clone https://github.com/beardedparrott/Orchicon.git
$ cd Orchicon
$ make build # → bin/orchicon
$ make dev-start # full dev environment
Prereqs: Go 1.26+, Node 22+, Docker. The Makefile installs buf and atlas via make tools.
Read the documentation
Comprehensive architecture, development, and user guide.