Status v0.1.13
Latest release: v0.1.13 (July 7, 2026)
| Phase |
Status |
Notes |
| MVP-0 Foundation |
✅ |
Compose, auto-migrate database, health, Prometheus/Grafana/Loki, CI |
| MVP-1 Script CRUD + Run |
✅ |
CRUD, run/stop, logs WS, import/export, templates |
| MVP-2 Scheduler + Dashboard |
✅ |
Cron/interval, KPI dashboard, webhooks |
| MVP-3 Editor + RBAC |
✅ |
Monaco editor, JWT auth, 4 roles, groups |
| Production-1 |
✅ |
Vault of secrets, notifications, backups |
| Production-2 |
✅ |
OTA updates, UpdateProvider, multi-runtime compose prod, metrics |
| Production-3 |
🔜 |
MQTT, HA Postgres, advanced isolation - backlog |
Phase overview
| Phase |
Goal |
Deadline (estimate) |
Result |
| MVP-0 |
Foundation |
2 weeks |
Compose up, health checks, empty UI shell |
| MVP-1 |
Script CRUD + Run |
3 weeks |
Creating/editing/running scripts, basic logs |
| MVP-2 |
Scheduler + Dashboard |
3 weeks |
Cron, KPI cards, history runs |
| MVP-3 |
Editor + Groups |
2 weeks |
Monaco multi-file, groups, basic RBAC |
| Production-1 |
Hardening |
4 weeks |
Secrets, notifications, backups |
| Production-2 |
Scale + OTA |
4 weeks |
Multi-runtime, OTA, full RBAC |
| Production-3 |
Enterprise |
ongoing |
MQTT, advanced isolation, HA Postgres |
MVP-0 – Foundation (weeks 1–2)
Goals
- Docker Compose stack starts with one command
- Backend connects to Postgres + Redis + MinIO
- Frontend renders the login page
- Runtime and Scheduler start and report on health
Results
- [x] Project structure and architecture documentation
- [ ]
docker compose up — all services healthy
- [ ] Alembic initial migration (users, scripts, runs)
- [ ]
GET /health, GET /api/v1/system/info
- [ ] Prometheus scrapes backend + runtime
/metrics
- [ ] Grafana with system dashboard template
- [ ] CI: lint + image build
Exit criteria
All containers undergo health check; The API returns 200 on /health.
MVP-1 - Script Management and Execution (Weeks 3-5)
Goals
- Full script life cycle without system restart
- Single-file scripts in sandbox with timeout and memory limit
Features
| Function |
Priority |
| Creating / editing / deleting a script |
P0 |
| Enable/disable script |
P0 |
| Manual run/stop |
P0 |
| Stream logs (WebSocket) |
P0 |
| History list runs |
P0 |
| Copying the script |
P1 |
| Import / export (zip) |
P1 |
| Basic templates (3 system) |
P1 |
Runtime (scope MVP)
- Subprocess + dedicated workspace
RLIMIT_CPU, RLIMIT_AS, wall-clock timeout- venv to the script from
requirements.txt
- Capture stdout/stderr → Redis stream → WS
API endpoints
POST /api/v1/scripts
GET /api/v1/scripts
GET /api/v1/scripts/{id}
PUT /api/v1/scripts/{id}
DELETE /api/v1/scripts/{id}
POST /api/v1/scripts/{id}/run
POST /api/v1/scripts/{id}/stop
GET /api/v1/scripts/{id}/runs
GET /api/v1/runs/{id}/logs
WS /ws/runs/{id}
Exit criteria
The user creates a script in the UI, runs it, sees logs and success/fail status.
MVP-2 - Scheduler and Monitoring (weeks 6–8)
Goals
- Cron and interval schedules
- Dashboard with KPI cards and basic graphs
Features
| Function |
Priority |
| Cron schedules |
P0 |
| Interval-schedules |
P0 |
| Max concurrent runs per script |
P0 |
| Window start/end date |
P1 |
| Webhook trigger (static token) |
P1 |
| KPI dashboard cards |
P0 |
| CPU/memory graphs (Grafana embed) |
P0 |
| Runs per day chart |
P1 |
| Errors in 24 hours |
P0 |
| Script status panel |
P0 |
Service scheduler
- APScheduler with Redis job store (shared state)
- Publishing to
runtime:jobs when triggered
- Reload by pub/sub
scheduler:reload
Exit criteria
The script is run via cron without manual action; dashboard shows live metrics.
MVP-3 - Editor, Groups and RBAC (Weeks 9–10)
Goals
- Production-level code editor
- Organizing scripts into groups
- Four roles with rights in the scope group
Features
| Function |
Priority |
| Monaco editor, Python syntax |
P0 |
| Multi-file project tree |
P0 |
Editor requirements.txt |
P0 |
| Syntax checking (ast.parse) |
P1 |
| Formatting (black, optional) |
P2 |
| Search by project |
P1 |
| CRUD groups with color/icon |
P0 |
| Roles: Admin, Developer, Operator, Viewer |
P0 |
| JWT auth + login page |
P0 |
Exit criteria
Developer edits a multi-file bot in the “bots” group; Operator can launch it, but not edit it.
Production-1 - Security, Secrets and Notices (Weeks 11–14)
Goals
- Vault of secrets, notification channels, backup/restore
Features
| Function |
Priority |
| Secrets for the script (encryption) |
P0 |
SDK platform.secrets.get() |
P0 |
| Email Notifications |
P1 |
| Telegram notifications |
P1 |
| Webhook notifications |
P0 |
| In-app notifications |
P0 |
| Manual backup |
P0 |
| Scheduled backup |
P1 |
| Restore from backup |
P0 |
| Export/import full configuration |
P1 |
| Audit log |
P1 |
| Run on script-complete event |
P1 |
| API trigger with auth |
P0 |
Exit criteria
Secrets are not visible in the code; backup restores scripts + database on a clean installation.
Production-2 - Scale, OTA and advanced runtime (weeks 15–18)
Goals
- Horizontal scaling runtime
- OTA update framework
- Detailed metrics on run
Features
| Function |
Priority |
| Multiple runtime replicas (Redis queue) |
P0 |
Interface UpdateProvider |
P0 |
Plug GitHubUpdateProvider |
P0 |
| OTA UI: check, apply, rollback |
P0 |
| Auto backup before update |
P0 |
| Sampling CPU/memory/threads on run |
P0 |
| cgroup v2 limits (Linux) |
P1 |
| Storage quotas per script |
P1 |
| API temp vs persistent storage |
P1 |
| MQTT client in SDK |
P2 |
| UI dependency tree |
P2 |
Exit criteria
2 runtime containers process a common queue; OTA apply + rollback checked for staging.
Production-3 - Enterprise and HA (ongoing)
Backlog
- Postgres / managed DB replication
- Network namespace isolation on sandbox
- SSO (OIDC/SAML)
- Multi-tenancy
- Custom Grafana dashboard per group
- Marketplace scripts/general templates
- Git sync script sources
- Canary runs / dry-run mode
- SLA alerting
Risk register
| Risk |
Mitigation |
| Slow venv build for many scripts |
Lazy build, venv cache, common core packages |
| Memory leak in user script |
Subprocess on run + hard memory cap + kill |
| Drift scheduler |
Redis distributed lock; one leader scheduler in MVP |
| Log volume |
Retention Loki; paginated API; optional S3 cold storage |
| Leaked secrets |
Do not return decrypted secrets via the API; access audit |
Matrix MVP vs Production
| Opportunity |
MVP |
Production |
| Script CRUD |
✅ |
✅ |
| Multi-file projects |
✅ (MVP-3) |
✅ |
| Sandbox insulation |
Basic rlimits |
+ cgroups, optional netns |
| Cron / interval |
✅ |
✅ |
| Event / chain triggers |
❌ |
✅ |
| Webhook trigger |
Basic |
✅ + rate limits |
| RBAC |
✅ (MVP-3) |
✅ + audit |
| Vault of secrets |
❌ |
✅ |
| Notifications |
❌ |
✅ |
| Backups |
❌ |
✅ |
| OTA updates |
❌ |
✅ |
| Multi-runtime |
❌ |
✅ |
| Prometheus/Grafana |
Basic |
Complete Dashboards |
| MQTT |
❌ |
✅ |
Recommended order of sprints (first 10 weeks)
Sprint 1: Compose, database migrations, health, auth foundation
Sprint 2: Script CRUD API + MinIO file storage
Sprint 3: Runtime sandbox v1 + manual run + WebSocket logs
Sprint 4: Frontend script list + run panel
Sprint 5: Scheduler cron + interval
Sprint 6: Dashboard KPIs + Grafana panels
Sprint 7: Monaco editor + file tree
Sprint 8: Groups + RBAC enforcement
Sprint 9: Import/export + templates
Sprint 10: Hardening, docs, MVP v0.1.0 release