Tool reference · Evaluation
memory_eval_entity_consistency
Checks that the same entity name is mapped to one canonical tag every time it is saved.
local stdio server read-only
When to use
- Tags for the same thing keep appearing in different spellings.
- After changing entity deduplication settings.
- You want a fast, offline pass/fail for tag canonicalisation.
Parameters
| Name | Type | Default | Description |
|---|---|---|---|
mode | "fast" | "balanced" | "deep" | "fast" | — |
Example
Arguments
{
"mode": "fast"
}Result shape
{
"scenarios_total": 3,
"scenarios_passed": 3,
"recall_at_5": 1,
"recall_at_10": 1,
"latency_ms": 0,
"mode": "fast",
"llm_calls_during_eval": 0,
"network_calls_during_eval": 0,
"attempts": [
{
"input": "Anthropic Claude",
"rewritten": [
"Anthropic Claude"
],
"decision": "merged"
}
]
}Runs on in-memory test data with a deterministic embedding; nothing is written to the store. Returns status "not_implemented" if the dedup module is missing.
Values are illustrative; the keys follow the server's handler. MCP clients receive the result as JSON text content.
Server description
The description the server sends to your agent in tools/list, captured from the v14.7.0 source:
v11.0 Phase 8: verifies entity_dedup canonicalization is stable across repeated saves of variant tag spellings.