3.8 KiB
3.8 KiB
Mission Control — Version History
v0.0.5 — 2026-05-12
- 7 monitoring summary endpoints implemented (Neo, Hudson security review)
- Cost summary endpoint (
/api/v1/monitoring/cost-summary) — consolidated cost overview per gateway with model breakdown - Cost breakdown endpoint (
/api/v1/monitoring/cost-breakdown) — models ranked by cost with percentage breakdown - Health summary endpoint (
/api/v1/monitoring/health-summary) — gateway health status and system metrics - Cron summary endpoint (
/api/v1/monitoring/cron-summary) — cron job status with scheduling info - Sessions summary endpoint (
/api/v1/monitoring/sessions-summary) — recent session events with context and token breakdown - Sub-agents summary endpoint (
/api/v1/monitoring/sub-agents-summary) — sub-agent runs with cost and token breakdown - Trends endpoint (
/api/v1/monitoring/trends) — cost and token trends over 7d or 30d time ranges - All endpoints support org-scoped filtering via
gateway_id, pagination, and security review passed (Hudson) - Backend builds and runs successfully with all Python files compiling cleanly
- Updated FUTURE.md — marked all 7 summary endpoints as DONE
v0.0.4 — 2026-05-11
- Phase 2: Gateway Data Collection Service
- Created
src/backend/app/services/monitoring/package with collector, Pydantic models, and init - GatewayCollectorService: background asyncio task polling gateway RPC endpoints
- Collection methods: cost (usage.cost + usage.status), cron (cron.list), sessions (sessions.list + preview), health (health + status)
- Upsert pattern for all DB writes to avoid duplicates
- Configurable intervals via env vars (COST=300s, CRON=60s, SESSION=30s, HEALTH=60s)
- Collector startup/shutdown integrated into main.py lifespan
- Fixed model FK definitions: replaced Column(UUID, ForeignKey(...)) with Field(foreign_key=...)
- Added missing gateway_id field to AlertRule model
- Backend builds and runs successfully, collector logs correctly
v0.0.3 — 2026-05-11
- Phase 2: Monitoring database models
- Add 7 new SQLModel tables for monitoring data collection
- CostSnapshot: timestamped cost readings from gateway RPC
- CronJobStatus: cron job state, schedule, failure tracking
- SessionEvent: session lifecycle events with token counts
- SubAgentRun: sub-agent execution records
- SystemHealthMetric: CPU, RAM, swap, disk, gateway status
- AlertRule: configurable alert thresholds with cooldown
- AlertEvent: triggered alert instances with acknowledgment
- Security: ON DELETE CASCADE on all FK constraints, SET NULL on acknowledged_by
- Performance: composite indexes on (org_id, created_at) and (org_id, gateway_id)
- Two Alembic migrations: table creation + CASCADE/index follow-up
- Frontend port changed to 3037
v0.0.2 — 2026-05-10
- Phase 1 CRITICAL: Base platform forked and running locally
- Copied base platform (Python/FastAPI backend + Next.js frontend) into src/
- Adapted Dockerfile and compose.yml for src/ project layout
- Fixed webhook-worker scripts path
- Local dev auth mode: local (token-based)
- Ports: backend 8080, frontend 3080 (avoiding conflicts with Portainer/open-webui)
- All 4 services running: db, redis, backend, frontend, webhook-worker
- 97 API endpoints verified operational
- Database migrations auto-applied
- Git repo initialized on main branch
v0.0.1 — 2026-05-10
- Project initialization
- Created project structure (PROJECT.md, STRUCTURE.md, FUTURE.md, HISTORY.md, DEVELOPMENT_LOG.md, VERSION.md)
- Cloned source repos: openclaw-mission-control, openclaw-dashboard, openclaw-pixel-agents-dashboard
- Architecture plan documented (4 phases: foundation, tracking, visualization, integration)
Version format: MAJOR.MINOR.PATCH New feature/page = minor bump, perf/refactor/fix = patch bump Bishop owns version bumps after verification