AgentTwin 0.2.2 · discovery and hosted workflow execution · Node.js 22.12+
What works today
Updated October 6, 2026. The public CLI is
agenttwin@0.2.2, with Node.js 22.12+ required. The website dashboard and authenticated API are deployed on Coolify. Later dashboard refinements are deployed to both dashboard URLs; the npm 0.2.2 package has not been republished for those UI-only changes.| Area | Available behavior |
| Discovery | Static project signals, workflow candidates, optional installed OpenCode session and a task for other coding agents |
| Account sync | Reviewed discovery metadata via authenticated, account-scoped npx agenttwin sync |
| Preparation | Explicit application files, invocation, operation mappings, fixtures, faults, assertions and source/spec digests |
| Execution | Reviewed bundled Node code in a fresh hosted E2B sandbox; no local application containers |
| Service behavior | Custom JSON collections and bounded Slack message/channel/history operations |
| Infrastructure | Real PostgreSQL 15 and Redis, with configured final-state checks |
| Evidence | Assertions, state, service events, application logs, sandbox metadata, comparison and exported fix tasks |
| Dashboard | Real imported/synced resources, movable dotted map, searchable workflows, jobs, cancellation, account menu and saved browser preferences |
Remaining gaps
Arbitrary vendor API compatibility, real third-party account provisioning, browser/macOS execution, universal workflow extraction, unattended repairs, persistent workflow definitions and automatic CI regression export are not implemented. Coding-agent analysis still needs reviewed invocation, fixtures and outcome expectations.
The API uses a single-instance durable disk store. Distributed scheduling, billing, deletion reconciliation and complete production monitoring remain work to do.
Verification and data
Engineering checks passed hosted execution, configured PostgreSQL/Redis assertions, account isolation/revocation, cancellation and persistence. The latest CLI suite passed 77 tests. Dashboard checks verified saved preferences, account connect/disconnect and full-width sidebar collapse on desktop and mobile. These checks are not customer adoption evidence or proof of production equivalence.
The production dashboard starts with an empty inventory and receipt list. The separate prototype uses fictional scenarios; its outcomes are scripted.