Architecture decision framework for evaluating whether Temporal Nexus is right for your use case. Use when user asks "should I use nexus", "nexus vs child workflows", "nexus vs activities", "when to use nexus", "cross-namespace pattern", "multi-namespace architecture", "nexus tradeoffs", or "nexus benefits". Covers tradeoff analysis, complexity scoring, and migration paths. Do NOT use for Nexus implementation details or code examples — use nexus-operations instead.
From timelordnpx claudepluginhub therealbill/mynet --plugin timelordThis skill uses the workspace's default tool permissions.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
Guidance for deciding whether Temporal Nexus is the right communication pattern for your architecture.
If all your workflows run in a single namespace, stop here — you don't need Nexus.
| Reason | Nexus Fit |
|---|---|
| Team ownership boundaries | Strong — Nexus provides clean service contracts |
| Security/compliance isolation | Strong — namespaces + Nexus maintain isolation |
| Independent scaling | Strong — separate workers, separate scaling |
| Environment separation (dev/staging/prod) | Not applicable — use same namespace structure per env |
| Just organizational preference | Weak — consider if the operational overhead is worth it |
| Need | Best Pattern |
|---|---|
| Fire-and-forget notification | Signals (if same NS) or Activities calling Signal API |
| Quick data lookup (< 10s) | Nexus sync operation |
| Long-running cross-team process | Nexus async (workflow-backed) operation |
| Cross-namespace, no durability needed | Activity with Temporal client (simpler, less durable) |
| Cross-namespace with full durability | Nexus (the primary use case) |
Nexus adds:
Nexus provides:
The question: Does the isolation and durability benefit outweigh the operational complexity for your case?
Multiple teams each owning a namespace, connected via Nexus endpoints:
┌──────────────┐ Nexus: payments-ep ┌──────────────┐
│ orders-ns │──────────────────────>│ payments-ns │
│ (Team A) │ │ (Team B) │
│ │ Nexus: inventory-ep ┌──────────────┐
│ │──────────────────────>│ inventory-ns │
└──────────────┘ │ (Team C) │
└──────────────┘
Good when: Teams need independence, clear ownership, separate deployment cycles.
Common platform service (e.g., notifications, payments) exposed via Nexus to all consuming namespaces:
┌────────────┐
│ service-a │──┐
└────────────┘ │ Nexus: notify-ep ┌────────────────┐
├────────────────────>│ notifications │
┌────────────┐ │ │ (platform team)│
│ service-b │──┘ └────────────────┘
└────────────┘
Good when: A central capability is consumed by many teams.
Don't use Nexus if caller and handler are in the same namespace. Child workflows give you the same decomposition with less overhead.
If you just need to call an HTTP API and don't need cross-namespace durability, use a regular Activity. Nexus is for Temporal-to-Temporal communication.
Rate your scenario:
| Factor | Low (1) | Medium (2) | High (3) |
|---|---|---|---|
| Number of consuming namespaces | 1 | 2-3 | 4+ |
| Handler operation complexity | Simple sync | Mix of sync/async | Complex chains |
| Team independence requirement | Nice to have | Important | Critical |
| Existing namespace topology | Single NS | Few NS, ad-hoc | Multi-NS established |
Score 4-6: Consider if simpler patterns (child workflows, activities) suffice. Score 7-9: Nexus is likely a good fit. Score 10-12: Nexus is strongly recommended.
If you're currently using ad-hoc cross-namespace communication (activities calling Temporal client, etc.):