Agent skill

Sanctions Screening

by lawve-ai in lawve-ai/awesome-legal-skills

Screen a client, counterparty, payer or corporate against the sanctions lists published by the designating authorities themselves — the UK Sanctions List (FCDO), OFSI, the UN Security Council, the…

MITAuto-check passed

Install Sanctions Screening

skills CLI
$ npx skills add lawve-ai/awesome-legal-skills --skill sanctions-screening -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install lawve-ai/awesome-legal-skills sanctions-screening --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/lawve-ai/awesome-legal-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/sanctions-screening-matei-clej .claude/skills/sanctions-screening && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
sanctions-screening
GitHub stars
847
Token cost
~3.3k tokens
SKILL.md length
1,668 words
Files
15
Skills in repo
154
Repo updated
First seen
Licence
MIT

At a glance

Screen a client, counterparty, payer or corporate against the sanctions lists published by the designating authorities themselves — the UK Sanctions List (FCDO), OFSI, the UN Security Council, the…

  • Works in 7 steps: A candidate is not a match. The score… → A nil return is bounded, and the record… → A failed read never reads as a clean… → …
  • Before taking a payment
  • SKILL.md covers Running it, The completeness gate, The lists and What the law requires, and of…, plus 3 more sections
  • Runs Python scripts from its folder; calls python3

What it does

Sanctions Screening is an agent skill from lawve-ai/awesome-legal-skills. Screen a client, counterparty, payer or corporate against the sanctions lists published by the designating authorities themselves — the UK Sanctions List (FCDO), OFSI, the UN Security Council, the EU consolidated list and OFAC's SDN and non-SDN lists. Use when onboarding, before taking a payment, before completing a transaction, or whenever a name must be checked against sanctions. Produces candidates and provenance, never a clearance: the record names every list searched with its publication date and count…

Its SKILL.md is about 3.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 16 other files (for example `README.md`, `TESTING.md` and `matching.py`).

The repository describes itself as: A curated list of awesome Agent Skills for automating legal work. The licence is MIT.

When your agent uses it

  • Before taking a payment
  • Before completing a transaction
  • Whenever a name must be checked against sanctions

Example prompts

  • “/sanctions-screening”

Requirements

  • Python 3

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

  1. A candidate is not a match. The score measures name similarity, nothing
  2. A nil return is bounded, and the record says by what — the lists actually
  3. A failed read never reads as a clean result. A list that is missing,
  4. Search more than one spelling of a transliterated name. The tool itself
  5. Screen the payer, not only the client. Third-party payers, corporate
  6. Re-screen. Designations are made weekly. A screen done at intake says
  7. Never complete the decision block on the fee earner's behalf, and never

What it can do on your machine

Read from SKILL.md and the folder at commit 045f738. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships script files (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Sanctions Screening loads about 3.3k tokens when it runs. Until then it costs about 240 tokens; SKILL.md has 1,668 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~240
When it runs · the whole SKILL.md, loaded when a task matches
~3.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from lawve-ai/awesome-legal-skills at commit 045f738, republished under its MIT licence (© lawve-ai). 1,668 words, ~3,278 tokens.

Download SKILL.mdSave it as .claude/skills/sanctions-screening/SKILL.md (or your agent's skills folder). This skill also uses 14 other files; get the full folder from GitHub.
name
sanctions-screening
description
Screen a client, counterparty, payer or corporate against the sanctions lists published by the designating authorities themselves — the UK Sanctions List (FCDO), OFSI, the UN Security Council, the EU consolidated list and OFAC's SDN and non-SDN lists. Use when onboarding, before taking a payment, before completing a transaction, or whenever a name must be checked against sanctions. Produces candidates and provenance, never a clearance: the record names every list searched with its publication date and count, states what was NOT searched, and ends at a blank decision block. Fails closed — a cache that is missing, damaged, truncated or stale is reported NOT SEARCHED and the process exits non-zero, so nothing can read an incomplete run as clean. Matches across transliteration, diacritics, homoglyphs and initials, because nobody spells a name the way the publisher does. Standard library only; screening opens no network connection.

Open-source sanctions screening

Three Python files, standard library only. No install step, no API key, no account, and no third-party screening vendor in the path — so there is no per-search cost and no reason to leave anybody unscreened.

The tool produces candidates and provenance. It never clears anybody. The comparison of a candidate against what is actually known of the subject, the decision that follows, and the record of that decision belong to the fee earner.

Running it

bash
python3 screen.py refresh                       # fetch + parse the default six lists (~30s)
python3 screen.py status                        # freshness, designation counts, failures
python3 screen.py sources                       # the registry, coverage and licences

python3 screen.py check "Ivan Petrov" --dob 1975-03-02 --nationality Russia
python3 screen.py check "Acme Trading LLC" --type entity --threshold 80
python3 screen.py check "Ramzan Kadyrov" --dob 05/10/1976 \
    --client MATTER-2026-0099 --matter "immigration advice" \
    --report ./screening/kadyrov-2026-09-01.md

python3 screen.py batch subjects.csv --report-dir ./screening/   # name,dob,nationality,type,client,matter

Exit codes: 0 complete search and nothing found, 2 candidates, 3 the search cannot be relied on, 4 the request was refused.

Switches: --source uk-sanctions-list,ofsi (restrict, or reach the non-default lists), --threshold (default 88 — see below), --include-delisted, --max-age-hours (default 24), --strict, --json, --verbose.

The default threshold is set from measurement. Recall is identical at every cut-off from 85 to 92 (99.7%) while false positives fall from 20% to 2%, and no true match scored between 85 and 92. Lower it to 82 to catch a query as thin as an initial plus one name, at a 40% false-positive rate; raise it to 92 for a first pass over a large book of clients.

Requires Python 3.11 or later (3.9 parses but is not tested). Nothing else. The cache lives in ~/.claude/sanctions-data (override with SANCTIONS_DATA) and is about 100 MB for the eight government lists; a run holds roughly 230 MB in memory while scanning.

The completeness gate

Every route out of this tool — text, JSON, a written record, a batch summary, an exit code — is rendered from one Outcome object, and an Outcome knows whether the search was complete. There is no path that reports "nothing found" without also carrying whether anything was actually searched.

A search is complete only if the request was valid and every list asked for was read in full, matched the digest recorded when it was fetched, and is current. Anything else is incomplete, and an incomplete search never exits 0.

ExitMeaning
0Complete search, no candidate at or above the threshold
2Candidates to consider
3The search cannot be relied on — a list was missing, damaged, altered, short, stale or never fetched
4The request itself was refused — an unusable option, a name with nothing searchable in it, a CSV with no name column

The cache is bound to its metadata by digest and parser version, so a list that has been altered, truncated or parsed by a different version of this code is refused rather than searched. Upgrading the tool therefore requires a refresh — that is the gate working, not a fault.

The lists

idListAuthorityDefault
uk-sanctions-listUK Sanctions ListFCDO✓
ofsiConsolidated List of Financial Sanctions TargetsHM Treasury / OFSI✓
unSecurity Council Consolidated ListUnited Nations✓
euConsolidated financial sanctions listEuropean Commission✓
ofac-sdnSpecially Designated NationalsUS Treasury / OFAC✓
ofac-consConsolidated non-SDN (SSI, FSE, NS-PLC, CAPTA)US Treasury / OFAC✓
canadaSEMA / JVCFOA consolidated listGlobal Affairs Canada
swissSECO sanctions listSwiss Confederation
opensanctions-sanctions~90 sources, deduplicatedOpenSanctions (aggregator)
opensanctions-pepsPolitically exposed personsOpenSanctions (aggregator)

The UK Sanctions List and the OFSI list are not the same thing and neither subsumes the other: the FCDO list is the legal list and carries every measure imposed (travel ban, arms, trade, transport, director disqualification), while the OFSI list carries the financial-sanctions targets and the asset-freeze detail. Both are searched by default.

The OpenSanctions datasets are CC BY-NC 4.0. Screening fee-earning client work is commercial use. Those two sources refuse to refresh until SANCTIONS_OPENSANCTIONS_LICENCE is set — set it to noncommercial for genuinely non-commercial work, or to your commercial licence reference once one is held. They are also the only PEP source here.

Per-source detail, quirks and failure modes: reference/sources.md.

What the law requires, and of whom

Verified against the in-force text on legislation.gov.uk on 02.09.2026. This is a UK-facing tool; practitioners elsewhere should read the equivalents.

The prohibitions reach the work, whoever the client is. UK regimes are made under the Sanctions and Anti-Money Laundering Act 2018 s.1, and s.21(1) sets their reach: prohibitions may be imposed in relation to conduct in the United Kingdom or its territorial sea by any person, and to conduct elsewhere only where it is by a United Kingdom person — s.21(2)-(3): a UK national (a British citizen, British overseas territories citizen, British National (Overseas) or British Overseas citizen, a British subject under the British Nationality Act 1981, or a British protected person), or a body incorporated or constituted under the law of any part of the UK. Read the two limbs separately. Limb (a) catches everything done in this jurisdiction whoever does it, so it reaches work carried out here regardless of counsel's nationality. Limb (b) reaches conduct abroad only if the actor is a United Kingdom person: practising here does not by itself make a foreign national one, though a UK-incorporated entity is one whatever the nationality of those behind it. Dealing with a designated person's funds, or making funds or economic resources available to or for the benefit of one, is prohibited by the regime regulations themselves. There is no client-type threshold and no de minimis, so this catches a fee taken from a third-party payer as readily as a corporate retainer.

Much litigation practice sits outside the MLR customer due diligence duty. An "independent legal professional" is defined in reg 12(1) of the Money Laundering Regulations 2017 (SI 2017/692) by reference to participation in financial or real property transactions — buying and selling property or business entities, managing client money or assets, opening or managing accounts, organising company contributions, and creating or managing trusts and companies. The test is what the retainer involves, not what the practice area is called: ordinary criminal, extradition and immigration advocacy will not usually engage that limb, but a matter that turns on any of those transactions does, whatever it is filed under. Where it is engaged, reg 33(1)(d) and reg 35 make PEP status an enhanced due diligence trigger (a trigger, never a prohibition), and reg 40(2)(a) with reg 40(3) require the CDD material to be kept for five years — running from the end of the business relationship for a relationship (reg 40(3)(b)), but from completion of the transaction for an occasional transaction (reg 40(3)(a)), which is the limb a one-off retainer falls into. Reg 40(4) caps relationship records at ten years, and reg 40(5) requires the personal data to be deleted once the period expires unless another duty or a legal-proceedings ground keeps it. Retention is a duty in both directions: keeping a screening record indefinitely is its own breach.

Show full SKILL.md (575 more words)Show less

The reporting duty is wider than the CDD duty, and does reach a legal practice. Under the Russia (Sanctions) (EU Exit) Regulations 2019 (SI 2019/855), reg 71(1)(d)(ii) makes a firm or sole practitioner providing "legal or notarial services" by way of business a relevant firm — with no transaction limb — and reg 70(1) requires a relevant firm to inform the Treasury as soon as practicable where it knows or has reasonable cause to suspect that a person is a designated person or has breached a prohibition, and the knowledge or suspicion came to it in the course of its business. Verified for the Russia regime; other regimes follow the same drafting, but the regime in play must be checked rather than assumed.

Ownership and control are not screened by any name search. Under reg 7 of SI 2019/855 a non-individual is owned or controlled by a designated person where that person holds, directly or indirectly, more than 50% of the shares or voting rights, or the right to appoint or remove a majority of the board — or, on the second limb, where it is reasonable to expect that they could achieve the result that the company's affairs are conducted in accordance with their wishes. An unlisted company can therefore be caught by a listed shareholder. That has to be worked out from the ownership structure; no list holds it.

The discipline

  1. A candidate is not a match. The score measures name similarity, nothing else. Adopting or discounting a candidate is a human comparison against date of birth, nationality, address, identifiers and the statement of reasons.
  2. A nil return is bounded, and the record says by what — the lists actually searched, the date each was published, and the spelling used. It is not a statement that the subject is not sanctioned.
  3. A failed read never reads as a clean result. A list that is missing, damaged, truncated, or short of the count recorded when it was fetched is reported as NOT SEARCHED on the face of the screening record, its hits are discarded rather than half-reported, and the run exits 3.
  4. Search more than one spelling of a transliterated name. The tool itself romanises Cyrillic only — under four systems, each searched — and any Cyrillic letter in a name triggers that. Arabic, Chinese and every other script are NOT romanised: a query in them is refused, and the passport or publisher spelling must be searched in Latin script. OFAC's files are entirely ASCII, so a non-Latin name can never match there except through a romanisation. The matcher folds diacritics and homoglyphs but cannot invent a spelling nobody gave it. Search the passport spelling and the common alternative.
  5. Screen the payer, not only the client. Third-party payers, corporate parents, funders and the opponent in a matter where money will move.
  6. Re-screen. Designations are made weekly. A screen done at intake says nothing about the position when the fee is taken.
  7. Never complete the decision block on the fee earner's behalf, and never describe a subject as "cleared". Fill in the search; leave the decision.

Privacy

Screening is entirely local. refresh contacts the eight publishing authorities listed above and nothing else; check and batch open no socket at all, which is asserted by a test that replaces the socket layer with something that raises. The name of a client is never sent anywhere. Screening records are written only where you point them.

Tests

bash
python3 tests/test_screen.py        # 130  matching, dates, the gate, parser output
python3 tests/integrity.py          # 186  field mapping, refresh, tampering, licence gate
python3 tests/adversarial.py        # 374  hostile input, cache damage, injection, privacy
python3 tests/release_gate.py       #      every published name retrieves its designation — run before a tag
python3 tests/benchmark.py          #     recall and precision against the lists
python3 tests/prefilter_safety.py   #     measures whether the speed filter costs recall
python3 tests/threshold_sweep.py    #     where the default threshold belongs

© lawve-ai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 14 other files in skills/sanctions-screening-matei-clej of lawve-ai/awesome-legal-skills.

  • SKILL.md
  • LICENSE
  • README.md
  • TESTING.md
  • matching.py
  • reference/sources.md
  • screen.py
  • sources.py
  • tests/adversarial.py
  • tests/benchmark.py
  • tests/integrity.py
  • tests/prefilter_safety.py
  • tests/release_gate.py
  • tests/test_screen.py
  • tests/threshold_sweep.py

Open the folder on GitHubat commit 045f738

Compare with similar skills

Sanctions Screening next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Sanctions Screening compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Sanctions Screening this skilllawve-ai/awesome-legal-skills847—~3.3kAutomated safety check: PassMIT
Counterparty Channel Disciplineaffaan-m/ECC277k—~2.3kAutomated safety check: PassMIT
Screen Adverse Mediasickn33/agentic-awesome-skills47k1 repos~1.1kAutomated safety check: PassApache-2.0
Bio Crispr Screens Screen QcFreedomIntelligence/OpenClaw-Medical-Skills3.1k—~2.1kAutomated safety check: PassNone
Screen Recordinggithub/awesome-copilot40k—~2kAutomated safety check: PassMIT
macOS Screen Recordersickn33/agentic-awesome-skills47k1 repos~727Automated safety check: PassMIT

Similar skills

  • Per-channel strict prompts, mention gating, silent observation, and a communication autonomy policy for agents that sit in shared channels with external counterparties.

    277k GitHub stars~2.3k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Screen Adverse Media

    sickn33/agentic-awesome-skills

    Screen a person or organisation for adverse media coverage, PEP status, and sanctions exposure — corroboration-gated, returns "review" never "guilty".

    47k GitHub starsUsed in 1 repo~1.1k tokens
    SecurityAuto-check passed
  • Bio Crispr Screens Screen Qc

    FreedomIntelligence/OpenClaw-Medical-Skills

    Quality control for pooled CRISPR screens. An agent skill from FreedomIntelligence/OpenClaw-Medical-Skills.

    3.1k GitHub stars~2.1k tokensUpdated 2 mo ago
    Research & ScienceAuto-check passed
  • Screen Recording

    github/awesome-copilot

    Official

    Create annotated animated GIF demos and screen recordings for pull requests and documentation.

    40k GitHub stars~2k tokensUpdated 2 days ago
    DevelopmentAuto-check passed
  • macOS Screen Recorder

    sickn33/agentic-awesome-skills

    macOS screen recorder that captures the main display PLUS system audio via ScreenCaptureKit — no BlackHole/loopback driver, no sudo, just the standard Screen Recording permission.

    47k GitHub starsUsed in 1 repo~727 tokens
    Auto-check passed
  • Synthetic Screen Recording

    calesthio/OpenMontage

    Synthetic terminal-style screen recording guidance for Remotion TerminalScene.

    66k GitHub stars~2.5k tokensUpdated 7 days ago
    Media & CreativeAuto-check passed

More from lawve-ai/awesome-legal-skills

All 154 skills in this repo
  • Customs Trade Law Onur Kafkas

    lawve-ai/awesome-legal-skills

    U.S. An agent skill from lawve-ai/awesome-legal-skills.

    847 GitHub stars~4.1k tokensUpdated 8 days ago
    Auto-check passed
  • Eu Data Act Oliver Schmidt Prietz

    lawve-ai/awesome-legal-skills

    Practitioner skill for advising on EU Regulation 2023/2854 (Data Act).

    847 GitHub stars~3.9k tokensUpdated 8 days ago
    Auto-check passed
  • Litigation Deadline Calendar

    lawve-ai/awesome-legal-skills

    Calendar litigation and arbitration deadlines from a scheduling order.

    847 GitHub stars~4.3k tokensUpdated 8 days ago
    Auto-check passed
  • Outlook Emails Lawvable

    lawve-ai/awesome-legal-skills

    Read, search, and download emails and attachments from Microsoft Outlook via OAuth2.

    847 GitHub stars~672 tokensUpdated 8 days ago
    Auto-check passed
  • Ambiguity Report

    lawve-ai/awesome-legal-skills

    Turn an interpretive-ambiguity audit of a legal text — contract, statute, regulation, or judicial opinion — into a polished deliverable.

    847 GitHub stars~4.6k tokensUpdated 8 days ago
    Auto-check passed
  • Az Eu Website Privacy Audit

    lawve-ai/awesome-legal-skills

    Audits a website for compliance with Azerbaijan's Law on Personal Data No.

    847 GitHub stars~3.8k tokensUpdated 8 days ago
    Auto-check passed

Questions about Sanctions Screening

What does Sanctions Screening do?

Screen a client, counterparty, payer or corporate against the sanctions lists published by the designating authorities themselves — the UK Sanctions List (FCDO), OFSI, the UN Security Council, the…. Sanctions Screening is an agent skill from lawve-ai/awesome-legal-skills. Screen a client, counterparty, payer or corporate against the sanctions lists published by the designating authorities themselves — the UK Sanctions List (FCDO), OFSI, the UN Security Council, the EU consolidated list and OFAC's SDN and non-SDN lists.

When should I use Sanctions Screening?

Sanctions Screening fits situations like: before taking a payment; before completing a transaction; whenever a name must be checked against sanctions.

How do I install Sanctions Screening in Claude Code?

Run `npx skills add lawve-ai/awesome-legal-skills --skill sanctions-screening -a claude-code`. Or copy the skill folder (skills/sanctions-screening-matei-clej in lawve-ai/awesome-legal-skills) into .claude/skills/sanctions-screening in your project. Claude Code loads it when a task matches its description.

How do I install Sanctions Screening in Codex?

Run `npx skills add lawve-ai/awesome-legal-skills --skill sanctions-screening -a codex`. Or copy the skill folder (skills/sanctions-screening-matei-clej in lawve-ai/awesome-legal-skills) into .agents/skills/sanctions-screening in your project. Codex loads it when a task matches its description.

Can I use Sanctions Screening in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add lawve-ai/awesome-legal-skills --skill sanctions-screening -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sanctions-screening, .gemini/skills/sanctions-screening, .github/skills/sanctions-screening and .opencode/skills/sanctions-screening in your project.

What does Sanctions Screening need to run?

Going by SKILL.md and its folder, Sanctions Screening needs Python for the scripts in its folder and the command-line tools its instructions call (python3). Our summary lists: Python 3.

Does Sanctions Screening access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Sanctions Screening safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Sanctions Screening use?

Sanctions Screening is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Sanctions Screening use?

About 3.3k tokens (SKILL.md is roughly 13k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Sanctions Screening?

Skills that share tags, products or a category with Sanctions Screening: Counterparty Channel Discipline (affaan-m/ECC, 277k stars), Screen Adverse Media (sickn33/agentic-awesome-skills, 47k stars), Bio Crispr Screens Screen Qc (FreedomIntelligence/OpenClaw-Medical-Skills, 3.1k stars) and Screen Recording (github/awesome-copilot, 40k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Sanctions Screening?

lawve-ai (a GitHub organization) maintains it in lawve-ai/awesome-legal-skills, which has 847 GitHub stars. The repository holds 154 skills in this directory. The repository was last updated on October 2, 2026.

Source: lawve-ai/awesome-legal-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.