Skip to content

Releases: dragocz1995/orcasynth

v1.1.0

20 Jun 15:16
@dragocz1995 dragocz1995
v1.1.0
10fee24

Choose a tag to compare

View all tags
v1.1.0 Latest
Latest

Orcasynth v1.1.0

Control autonomous coding agents — without losing control.

Highlights since v1.0.1

  • Pilot & Overseer CLI agents — autopilot planning and oversight now run as real coding-agent sessions on a task-tree mission engine, with autonomy levels (L0–L3) and guardrails.
  • Token & cost observability — usage is read from each agent's local session storage, attributed per task, and shown live in the UI while an agent runs.
  • Settings auto-save — adding/editing/removing a model persists immediately (no separate save step).
  • Cleaner sessions — daemon-classified session identity (agent / pilot / overseer roles) and assorted web polish.
  • Docs overhaul — full REST API reference, architecture, concepts, and guides.

Install

See the README and docs/.

Full changelog: v1.0.1...v1.1.0

Assets 2
Loading

Orcasynth v1.0.1 — initial public release

20 Jun 06:57
@dragocz1995 dragocz1995
v1.0.1
646b09f

Choose a tag to compare

View all tags
Orcasynth v1.0.1 — initial public release

Orcasynth is a self-hosted daemon that turns a goal into a plan, runs coding agents (Claude Code, OpenCode, Codex) in isolated tmux sessions, and supervises them to completion. It ships a REST API, a CLI, and a real-time web UI.

This is the first public release.

Highlights

  • Autopilot planning — give the Pilot a goal; an LLM decomposes it into ordered phases, names an agent per phase, and chains them by dependency.
  • Agent-agnostic spawning — run Claude Code, OpenCode, or Codex in tmux, configurable per task. Each agent gets the task context and closes its own task when done.
  • Autonomy levels (L0–L3) — the overseer auto-clears safe permission prompts at higher autonomy and escalates destructive or uncertain ones to a human.
  • Live web UI — tasks, a kanban board + calendar, missions with progress, a timeline, a built-in Monaco editor, and live tmux session previews with one-click intervention. EN/CS i18n built in.
  • Guardrails & self-healing — sensitive work (schema, auth, payments, destructive ops) is blocked until cleared, with an optional LLM overseer gate; a stuck-session detector revives agents that die without closing out, and live token/cost usage is shown per run.
  • Multi-user RBAC — per-project assignments, per-user model allow-lists, profiles & avatars, and a first-run onboarding that needs no login until the first admin is created.
  • Self-hosted & lightweight — a single SQLite-backed daemon (Hono) + a Next.js front end. No external services required beyond your LLM provider.

Getting started

Requires Node ≥ 22 and tmux. See the README for the quick start.

Contributing

Contributors welcome — open a feature request, file a bug report, or send a PR. See CONTRIBUTING.md.

Loading