Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.linksee.app/llms.txt

Use this file to discover all available pages before exploring further.

Explicitly delete a memory by ID, or run auto-forgetting across all memories based on the Ebbinghaus forgetting curve. Caveat-layer, goal-layer, and pinned (importance >= 0.9) memories are always preserved.
forget is destructive. Prefer update_memory for corrections.

​
Parameters

​
memory_id
number
Specific memory to delete. If omitted, runs an auto-sweep.
​
dry_run
boolean
default:"false"
Report what would be deleted without actually deleting.
​
interactive
boolean
default:"false"
If true, ask the user to confirm via MCP Elicitation before deleting. Only applies when memory_id is set.

​
Auto-forgetting

When called without memory_id, runs an automatic sweep using the forgetting risk formula: 
risk = (1 - heat/100) * (1 - importance) * daysSinceAccess * (1 + daysSinceAccess/30) * altitudeMultiplier

​
Altitude multipliers

AltitudeMultiplierEffect
mission0Never forgotten
strategy0.1Very slow decay
architecture0.3Slow decay
implementation1.0Normal decay

​
Thresholds

Risk scoreAction
< 50Keep
50 - 200Compress (merge into learning)
> 200Drop

​
Always preserved

These memories are never auto-forgotten:
  • All caveat layer memories
  • All goal layer memories
  • All memories with importance >= 0.9 (pinned)
  • All memories with protected = true

​
Example: Dry run

{
  "dry_run": true
}
Returns a list of memories that would be deleted or compressed, without modifying anything.