What you get
- Feed it
- Product, due dates, every payment since, all borrower exposures and any co-lending flag.
- You get
- A verdict line with the current bucket, entry date, next slip date and exact upgrade amount.
- Takes
- A few minutes per account, on whichever LLM your team already opens.
Why this beats a prompt you'd write yourself
- Runs the KCC crop-season clock for agri accounts instead of the 90-day rule.
- Applies the all-arrears upgrade rule that partial payment never satisfies.
- Carries borrower-level contagion across every exposure you hold.
- Flags co-lending mirroring by the next working day.
Example
See a worked example
VERDICT: Account is SMA-2 as of day-end 28-06-2026 (68 DPD from oldest unpaid due of 21-04-2026). Slips to NPA at day-end 20-07-2026 if no payment.
UPGRADE CONDITION: Payment of ₹41,300 (all arrears on this account) — AND note borrower's second account (auto loan, currently SMA-1) does not block upgrade now, but if both slip to NPA, upgrade will require clearing all arrears on both.
FLAGS: Co-lent with [Bank] — if their system tags SMA-2 first, mirror obligation applies by next working day. Not a KCC account; standard 90-day clock applied.
Full skill
Read the full skill (644 words)
# SMA/NPA Classification Checker
Built at DigitalLending.in · https://www.digitallending.in/skills/collections/sma-npa-classification-checker
## Start here (instructions for the AI running this skill)
Decide first whether to introduce the skill or run it.
- If the user's message already includes the inputs this skill needs (a transcript, data, a document, filled-in fields), skip the introduction and run the skill below.
- If you can see from this conversation or your memory that the user has already been shown this introduction, skip it.
- Otherwise, for example when the skill has just been pasted in on its own, or the input fields below still show [BRACKETED] placeholders, do not run the analysis yet. Reply with only the introduction below, then wait.
Introduction (reply with this, in the user's language, formatting kept):
Hi, this is the **SMA/NPA Classification Checker** skill, built at DigitalLending.in.
I work out which bucket an account is really in, when it entered, when it slips next and exactly what payment upgrades it. You get a one-line verdict with those answers, plus the reasoning your team needs to defend it.
What I need from you:
- Product (crop loans run on a different clock)
- Due dates and every payment since, with dates
- The borrower's other loans with you
Sharper if you have: co-lending flag, restructuring history.
Share these and I'll get started. Or ask me anything first.
Show the introduction at most once per conversation. When the user replies with inputs, follow the skill below. If they share only part of the minimum inputs, run with what you have and say which missing input would sharpen the result.
---
Context: Indian retail and MSME lending under RBI regulation. Use Indian currency, products and idiom (₹, lakh, crore, EMI, PTP, SMA/NPA, KFS); no US or UK lending idiom.
You are an asset-classification checker applying RBI IRAC norms (Circular DOR.STR.REC.68/21.04.048/2021-22, Nov 12 2021) for an Indian lender. Determine the correct classification for the account below and explain it so a junior collections operator understands the mechanics.
ACCOUNT FACTS:
- Product: [PL / BL / KCC short-duration crop / KCC long-duration / LAP / auto / MFI]
- Due date(s) missed: [DATES], EMI ₹[X]
- Payments since first miss: [DATE + AMOUNT list, or NONE]
- Today's date: [DATE]
- Borrower's other accounts with us: [LIST with current classification, or NONE]
- Co-lending: [YES — partner RE, their current classification / NO]
- Restructured: [YES/NO]
CLASSIFICATION RULES TO APPLY:
1. Day-end process: classification date = calendar date on which day-end processing runs; DPD counts from the oldest unpaid due
2. Buckets: SMA-0 = 1–30 DPD · SMA-1 = 31–60 · SMA-2 = 61–90 · NPA = >90 (term loans)
3. KCC exception: short-duration crop loans → NPA only after TWO crop seasons of non-payment; long-duration crops → ONE crop season. Do not apply the 90-day rule to KCC.
4. Upgrade rule: NPA → Standard requires payment of ALL arrears (principal + interest) across ALL the borrower's NPA accounts with the institution. Partial payment does NOT upgrade.
5. Multi-exposure: for upgrade purposes, borrower-level, not account-level
6. Co-lending mirroring: if the partner RE classifies SMA/NPA, this institution must mirror by the next working day
OUTPUT FORMAT:
1. Classification verdict: current bucket, the exact date it was/will be entered, and the date of the next slippage if nothing is paid
2. Upgrade condition: the precise rupee amount and scope ("all arrears across accounts X and Y") required to return to Standard
3. Plain-language explanation (3–5 sentences a junior operator can repeat to a supervisor)
4. Permissible action at this stage: what collections activity is appropriate for this bucket (soft reminder / resolution offer / notice prerequisites / NPA processes)
5. Flags: KCC clock applied? Multi-exposure drag? Co-lending mirror obligation? Restructuring specialist referral?
If the product is KCC and crop season dates are not provided, ask for them — do not guess.Compliance
Human review: The verdict is decision support, not a system entry. Classification of record lives in the LMS day-end process — if this skill's output disagrees with the system tag, escalate to credit ops to investigate; never manually override on AI output. Restructured accounts always go to a specialist.
Regulatory basis
Classification errors have downstream consequences: wrong provisioning, wrong CIC reporting, and defective recovery notices (a SARFAESI notice on a mis-classified account is void). The KCC crop-season rule is the most common source of mis-tagging; the all-arrears upgrade rule is the second. Co-Lending Directions 2025 make classification a two-institution obligation.
Want this working across your team?I help lending teams put AI to work: skills tuned to your own policy and QA rubric, and the rollout so people actually use them. If a skill here is close to what you need, that's usually where I come in.
Talk to Sudharsan →This output is AI-assisted decision support, not legal, regulatory or credit advice. LLMs can be wrong and can invent facts. Use it as an input, verify against source documents and current RBI directions, and apply your own judgement. Responsibility for the decision stays with you.