Unclassified

Long-term Task Management MCP

Canonical design for Hermes's unified long-running and event-driven task execution layer.

1. Automation Tasks

Business workflows that run on a schedule.

  • Renewal risk scoring
  • Customer health recalculation
  • Opportunity pipeline refresh
  • Weekly executive summaries
  • SLA compliance checks
  • Invoice reconciliation
  • License utilization reports

2. Watch Tasks

Run only when something changes.

Watch GitHub releases
Watch Salesforce schema changes
Watch HubSpot property changes
Watch Cloudflare deployments
Watch production incidents
Watch documentation updates

These are condition-based rather than purely time-based.

3. Memory Tasks

Maintain the Adaptive Spine.

  • Memory compaction
  • Duplicate detection
  • Entity merging
  • Timeline rebuilding
  • Embedding regeneration
  • Knowledge graph updates
  • Continuity validation

4. AI Tasks

Background intelligence.

  • Overnight account summaries
  • Ticket clustering
  • Meeting summarization
  • Opportunity forecasting
  • Churn prediction
  • Next-best-action generation
  • Knowledge synthesis

5. Integration Tasks

Connector operations.

  • Full sync
  • Incremental sync
  • Webhook verification
  • OAuth validation
  • Connector health checks
  • Schema discovery
  • Rate-limit monitoring

6. Infrastructure Tasks

Platform maintenance.

  • Worker health
  • Queue depth monitoring
  • Cache warming
  • Database optimization
  • Secret rotation checks
  • Backup verification
  • Storage cleanup

7. Governance Tasks

Compliance and policy.

  • Approval reminders
  • Policy drift detection
  • Permission audits
  • Access reviews
  • Audit report generation
  • Compliance validation

8. Security Tasks

Continuous security monitoring.

  • Token expiration checks
  • Secret exposure detection
  • Failed login analysis
  • API key validation
  • Certificate expiration
  • Suspicious activity detection

9. Workspace Tasks

Per-workbench automation.

  • Refresh dashboards
  • Update KPIs
  • Refresh timelines
  • Compute priorities
  • Assign work
  • Generate daily briefings

10. Agent Tasks

Hermes self-management.

  • Evaluate completed plans
  • Retry failed executions
  • Clean execution history
  • Optimize workflows
  • Update capability cache
  • Learn from execution outcomes

11. Marketplace Tasks

Connector ecosystem.

  • Discover new connector versions
  • Validate installed connectors
  • Update connector metadata
  • Check marketplace health

12. Observability Tasks

Operational telemetry.

  • Aggregate metrics
  • Trace analysis
  • Error clustering
  • Latency analysis
  • Cost reporting
  • Capacity forecasting

Long-term Task Management MCP

I would add another internal service to the platform:

Hermes
    │
    ▼
Task Management MCP
    │
    ├── Scheduler
    ├── Event Watches
    ├── Workflow Engine
    ├── Retry Manager
    ├── Notifications
    ├── History
    ├── Dead Letter Queue
    └── Metrics

Core capabilities:

tasks.create
tasks.schedule
tasks.watch
tasks.run
tasks.pause
tasks.resume
tasks.cancel
tasks.retry
tasks.history
tasks.logs
tasks.metrics

Two execution models

Rather than relying only on cron, support both:

  1. Time-based

    • Every hour
    • Every day
    • Every Monday
    • First day of the month

  2. Event-based

    • GitHub push received
    • Salesforce opportunity updated
    • Slack message posted
    • New customer onboarded
    • OAuth connection revoked
    • Cloudflare deployment completed

This lets Hermes respond to real operational events while still handling recurring maintenance, giving you a unified execution model for automation across the platform.