CC sanctions-screening
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.
Screen a client, counterparty, payer or corporate against the sanctions lists published by the designating authorities themselves — the UK Sanctions List…
As a process C 50/100 · Has gaps — weak spots: result and completion, inputs and preconditions, consistency
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The files contain invisible characters, encoded commands or comments hidden from readers but visible to the model. What you read differs from what the agent sees.
Remove invisible characters (they usually sneak in through copy-paste) and encoded strings: no catalog will pass them. Instructions for the model must be readable by a human too.
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 4
✓ No critical or high findings
Medium and low: 4
-
medium Obfuscation
uni-mixed-script-wordmatching.py:60Word mixing Latin and Cyrillic letters (homoglyph obfuscation) (1 occurrence)# otherwise-Latin names — "ROMASHKІN" on the EU list carries a Cyrillic І — and
-
medium Obfuscation
uni-mixed-script-wordreference/sources.md:68Word mixing Latin and Cyrillic letters (homoglyph obfuscation) (1 occurrence)* **Cyrillic homoglyphs appear inside otherwise-Latin names** — `ROMASHKІN`
-
medium Obfuscation
uni-mixed-script-wordsources.py:334Word mixing Latin and Cyrillic letters (homoglyph obfuscation) (1 occurrence)Latin homoglyph (Ольга Валерiївна — that "i" is Latin). Those were landing
-
low Obfuscation
uni-mixed-script-wordTESTING.md:158Word mixing Latin and Cyrillic letters (homoglyph obfuscation) (1 occurrence) (test fixture / example file)- **Cyrillic homoglyphs split surnames.** The EU list publishes `ROMASHKІN` with
fixture
Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 50/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 5 mutating operations with no state check
- 40Consistency. Frontmatter name (sanctions-screening) differs from the folder (sanctions-screening-matei-clej)
- 55Failures and branches. 1 branches
- 60Tools and files. Uses tools (web, python) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 100Steps. 7 steps
- 100Execution cost. Instruction body is 3031 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 940: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +4Structure: 8 headings
- +3Step-by-step instructions: 7 items
- +4Has examples (2 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 86.