agentleFS
Sign inStart free
NVIDIAKnown publisherNVIDIA/nemoclaw-community/examples/recipes/nvidia/payment-ops-hermes/skills/payment-screening/SKILL.md

payment-screening

Skill142 starsChanged 3 months agoApache-2.0View on GitHub

Screen an outbound payment against limits, sanctions/watchlists, and duplicate checks before a human releases it.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when an operator asks you to check, validate, or screen an outbound payment (wire or ACH) before release.

    line 8
  2. Very specific:

    Run the screener against a payment ID from the queue (omit --id to screen the whole batch):

    line 28
  3. Distinctive:

    Never recommend overriding a hold.

    line 54
  4. Distinctive:

    Do not claim a payment passed if any check failed.

    line 59
  5. Distinctive:

    Do not invent a sanctions match or clear a real one — report what the OFAC SDN fixture returned, with the matched name and program.

    line 60
  6. Distinctive:

    Do not release the payment.

    line 62

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

payment-screening

Use this skill when an operator asks you to check, validate, or screen an outbound payment (wire or ACH) before release.

This skill is read-only. It does not release, send, or settle anything. Release requires a human approver — see the release-packet skill.

Inputs

  • Queue: an ISO 20022 pain.001 credit-transfer file (data/payment-queue.json). The screener flattens every CdtTrfTxInf transaction and screens it by EndToEndId.
  • Sanctions: a bundled, dated OFAC SDN public-data fixture (data/ofac-sdn-fixture.json). It contains three real public SDN records selected for deterministic demonstration; it is not a complete sanctions list. Only the payments are synthetic.

Procedure

1. Screen the payment

Run the screener against a payment ID from the queue (omit --id to screen the whole batch):

/usr/bin/python3 /sandbox/.hermes/skills/payment-screening/scripts/screen_payment.py \
  --queue /sandbox/.hermes/skills/payment-screening/data/payment-queue.json \
  --sanctions /sandbox/.hermes/skills/payment-screening/data/ofac-sdn-fixture.json \
  --id WIRE-1007

The screener returns structured JSON with a decision (CLEARED_FOR_REVIEW or HOLD) and a per-check breakdown. Sanctions hits cite the matched OFAC SDN name and program.

2. Report the result

Use this structure:

  • Decision: CLEARED_FOR_REVIEW or HOLD
  • Checks: limit · sanctions · duplicate · beneficiary (pass/fail each)
  • Exceptions: the exact rule that fired, in plain language
  • Next step: what the human checker should verify before releasing
  • Caveat: screening is support; the human approver is accountable

3. On HOLD

Never recommend overriding a hold. State the rule, and route to the appropriate human queue (sanctions analyst, treasury, compliance).

Pitfalls

  • Do not claim a payment passed if any check failed.
  • Do not invent a sanctions match or clear a real one — report what the OFAC SDN fixture returned, with the matched name and program.
  • Do not release the payment. You cannot reach the payment rail, and you must not imply that you can.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free