Back to catalog / Developer tools
Event store design: preserve append-only history and rebuildable projections
Plan event streams, concurrent writes, global ordering, and replayable read models.
Skills
Define aggregate streams and event schemas with unique IDs, aggregate sequence, occurrence time, and version. Use append-only writes with optimistic concurrency; never overwrite prior events. Decide whether a global position is required instead of treating timestamps as a total order. Make projection consumers idempotent and rebuildable from checkpoints; specify snapshots, replay boundaries, and schema upcasting. Compare retention, archival, tiering, and query indexes before choosing storage. Test replay consistency and duration on an isolated copy before production replay.
Permissions
Prerequisites
- For event-store architecture, streams, global ordering, projections, or scaling.
- Requires aggregate/stream boundaries, event types, concurrency, and retention/replay needs.
Permissions
- May change storage schemas, event writers, and projection consumers.
- Events are immutable appended facts; replay/migration affects downstream state and must be tested in isolation.
Limitations
Enforce event identity, aggregate sequence, and idempotency; projections are rebuildable, event mutation breaks history.
Pinned upstream source
- Repository
- wshobson/agents
- File path
- plugins/backend-development/skills/event-store-design/SKILL.md
- Pinned commit
46891e7e60da0e52baf1050b7b6391b64e84c6d9- Source checked
- 2026-10-14 · Reviewed against the pinned upstream source
Repository metrics
These values describe the upstream repository, not popularity or likes for an individual skill or playbook.
- Stars
- Not collected
- Forks
- Not collected
- Observed at
- Not collected
No metrics snapshot is available yet.
License: MIT
Compatible tools
Event store schema · Append-only log · Projection consumers
Main topic
Take a coding Agent change from context to verification