logs

This DB is for APPLICATION LEVEL AUDIT LOGS. It will not capture any low level errors, but will instead track events and provide traceability for things like ingestion, enrichment, and tagging. Could support user-facing history / admin dashboards as well.

Use Case	Example
Debugging	“Why didn’t this import result in new memories?”
Auditing	“Which prompt was used in this ingestion?”
UI Feedback	“Your import succeeded with 15 entries.”
Developer Insight	“Which subsystems are getting hit?”
NOTE: Best Practices for App-Level logs in MongoDB

---

Schema

{
  "_id": ObjectId,
  "event_type": "import_processed",     // or 'error', 'memory_tagged'
  "user_id": ObjectId,                  // who initiated the action
  "import_id": ObjectId,                // optional, related import
  "memory_id": ObjectId,                // optional, related memory
  "prompt_id": ObjectId,                // optional, related prompt
 
  "details": {
    "message": "15 memories ingested, 2 duplicates skipped.",
    "tagging_mode": "auto",
    "trigger": "manual"
  },
 
  "level": "info",                      // info | warning | error
  "created_at": ISODate
}

Ideas for Later

Field	Purpose
request_id	Attach logs to a traceable external request or workflow
subsystem	e.g. “echoforge”, “contextweave”, “akasha”
error_stack	If level === error, capture traceback