UK Due Diligence

Overview Schema Source Related Servers Discussions Admin

Screen a Name Against Sanctions Lists

sanctions_screen

Read-onlyIdempotent

Screen a person or company name against UK, US, EU and UN sanctions lists and return matching entries with regime, source reference and listing date.

Instructions

Screen a name against the UK (OFSI), US (OFAC), EU and UN consolidated sanctions lists.

Returns every list entry whose primary name or alias matches, with the regime, source reference and listing date where the source gives one. Use it to check whether a counterparty — or its officers / persons with significant control — appears on a sanctions list.

MATCHING is deterministic: normalised exact + alias match (case-, accent- and punctuation-insensitive). A company/entity legal name matches reliably; PERSON names with transliteration variants may not (e.g. 'Mohammed' vs 'Muhamad'). An empty result is therefore NOT a guarantee of clearance, and a hit on a common name may be a false positive to disambiguate. This is a screening aid, not a compliance determination.

lists_screened reports which of OFSI/OFAC/EU/UN were actually loaded; lists_unavailable names those that were not, and is_partial is true whenever it is non-empty or the lists are a previous load that could not be refreshed (as_at then shows when they were loaded). When is_partial is true an empty hits is UNRESOLVED, not clearance. If no list could be loaded at all the call raises a retryable error rather than returning an empty screen. as_at is when the lists were last refreshed on this server.

Input Schema

Name	Required	Description	Default
`name`	Yes	Person or company/entity name to screen against the consolidated sanctions lists.
`entity_type`	No	Optional filter: 'person' or 'entity'. Omit to screen both.

Output Schema

Name	Required	Description
`hits`	No	Matching list entries. An empty list means no exact/alias match on the screened lists — NOT a guarantee of clearance (see the tool description on matching limits).
`as_at`	No	When this server last refreshed the loaded lists (ISO timestamp). Provenance for the screen — the lists update on designation.
`query`	Yes	The name that was screened.
`is_partial`	No	True when any expected list was unavailable, or when the lists are a previous load that could not be refreshed (see as_at). When true, an empty hits list must be treated as UNRESOLVED, not as clearance.
`match_count`	Yes	Number of list entries that matched the query.
`lists_screened`	No	Which consolidated lists were fully loaded and actually screened for this call. A list absent here was NOT screened — see lists_unavailable. Note a list that failed part-way still contributes the entries it had already parsed, so a hit's list_source may name a list missing from here; that is a real match on a list this server does not consider fully screened, not a contradiction.
`normalized_query`	Yes	The normalised form used for matching (upper-cased, accent- and punctuation-stripped, whitespace-collapsed).
`lists_unavailable`	No	Expected lists that did not load, or loaded to zero records, and so were NOT screened. Non-empty means this screen is incomplete: a name on one of these lists would not appear in hits. Absence of evidence here is not evidence of absence.
`entity_type_filter`	No	entity_type filter applied to the screen ('person'/'entity'), or null.

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedv1.6.0Oct 29, 2026

changedOutput schema / properties / hits / items / properties / reference / description

Previous value: -"The source's own reference for the listing (OFSI Group ID, OFAC uid, EU reference number, or UN reference number)."New value: +"The source's own reference for the listing (FCDO Unique ID, OFAC uid, EU reference number, or UN reference number)."

changedOutput schema / properties / is_partial / description

Previous value: -"True when any expected list was unavailable. When true, an empty hits list must be treated as UNRESOLVED, not as clearance."New value: +"True when any expected list was unavailable, or when the lists are a previous load that could not be refreshed (see as_at). When true, an empty hits list must be treated as UNRESOLVED, not as clearance."

Changed3 schema fields changedv1.4.0Sep 12, 2026

addedOutput schema / properties / is_partial

Added value: +{
+  "default": false,
+  "description": "True when any expected list was unavailable. When true, an empty hits list must be treated as UNRESOLVED, not as clearance.",
+  "type": "boolean"
+}

changedOutput schema / properties / lists_screened / description

Previous value: -"Which consolidated lists were loaded and actually screened for this call. A list absent here failed to load and was NOT screened — treat the result as partial if any of OFSI/OFAC/EU/UN is missing."New value: +"Which consolidated lists were fully loaded and actually screened for this call. A list absent here was NOT screened — see lists_unavailable. Note a list that failed part-way still contributes the entries it had already parsed, so a hit's list_source may name a list missing from here; that is a real match on a list this server does not consider fully screened, not a contradiction."

addedOutput schema / properties / lists_unavailable

Added value: +{
+  "description": "Expected lists that did not load, or loaded to zero records, and so were NOT screened. Non-empty means this screen is incomplete: a name on one of these lists would not appear in hits. Absence of evidence here is not evidence of absence.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

Addedv1.1.1Jul 29, 2026

TDQS

A4.6/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the readOnly/idempotent annotations: it discloses the deterministic normalised exact+alias matching rule, warns that person-name transliteration variants produce false negatives and common names produce false positives, and explains that an empty result is not clearance. It also documents partial-load semantics (`lists_screened`, `lists_unavailable`, `is_partial`, `as_at`) and the retryable error raised when no list loads.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Long but front-loaded and well structured: purpose first, then matching caveats, then partial-list/error semantics. Each paragraph carries operational weight (false positive/negative risk, partial data handling), though the final `as_at` sentence is somewhat repetitive with the earlier partial-load explanation.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, yet the description still explains the key response fields that determine whether an empty result can be trusted, which is the highest-risk ambiguity for this tool. Combined with the error behavior and matching algorithm, an agent has everything needed to call and interpret it correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds real meaning by contrasting entity legal names (reliable) with person names (transliteration risk), which is exactly the rationale for the optional `entity_type` filter. It does not restate the min/max length constraints, but those are already in the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('screen a name against the UK (OFSI), US (OFAC), EU and UN consolidated sanctions lists') and enumerates exactly which regimes are covered. No sibling tool does sanctions screening, so it is unmistakable among the listed alternatives.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly says 'Use it to check whether a counterparty — or its officers / persons with significant control — appears on a sanctions list', which gives a clear when-to-use context. It also draws the boundary 'This is a screening aid, not a compliance determination', though there is no sibling alternative to route away from, so no exclusion clause is needed.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server