Suggested match
Explanation
Generated from the read-only evidence below. It never changes a match or a record.
The bank narration contains a truncated UTR that almost matches the settlement UTR, causing a fuzzy match suggestion.
The engine identified that the narration token "AXISN26294239" is a truncated version of the settlement UTR "AXISN26294239504". The amount (₹65,949.06) and transaction date (2026-10-21) are identical, giving the suggestion a confidence of 0.8125. Because the UTR is not an exact match, a human needs to verify whether it is a typographical omission or a different transaction.
Check the UTR on the original bank statement; if it is a typo confirming the same transaction, approve the match, otherwise reject it.
Evidence records
Read-only source records involved in this exception.
- txn date
- 2026-10-21
- narration
- NEFT CR-AXISN26294239-GATEWAY SETTLEMENTS PVT LTD-SETTL MIDG3NF3BFV2PY
- credit
- ₹65,949.06
- debit
- ₹0.00
- balance
- ₹1,55,03,098.96
- utr
- AXISN26294239504
- amount
- ₹65,949.06
- settled at
- 2026-10-21
- created at
- 2026-10-21T09:05:00+05:30
- item count
- 17
- fee total
- ₹1,178.45
- gst total
- ₹212.15
Rule trace
The deterministic rule that fired and what it saw.
{
"rule": "stage2.fuzzy_suggest",
"reason": "truncated_utr",
"best": {
"settlement_id": "setl_9kAIAxmmHHpVt3",
"utr": "AXISN26294239504",
"token": "AXISN26294239",
"token_source": "narration",
"distance": null,
"truncated": true,
"confidence": 0.8125,
"amount_equal": true
},
"runner_up": null,
"margin": 0.8125,
"suggested": {
"settlement_id": "setl_9kAIAxmmHHpVt3",
"utr": "AXISN26294239504",
"token": "AXISN26294239",
"token_source": "narration",
"distance": null,
"truncated": true,
"confidence": 0.8125,
"amount_equal": true
},
"candidates": [
{
"settlement_id": "setl_9kAIAxmmHHpVt3",
"utr": "AXISN26294239504",
"token": "AXISN26294239",
"token_source": "narration",
"distance": null,
"truncated": true,
"confidence": 0.8125,
"amount_equal": true
}
]
}Audit history
Append-only. Actor: system, llm or human.
- llminvestigation
source=llm · model=openai/gpt-oss-120b · review_policy_applied=false
- systemjournal draft
applicable=false · reason=SUGGESTED_MATCH needs a matching decision, not a journal entry