The corpus activity feed — recent ownership-change and filed-accounts events, cited. Pass since= to stream only new events. The default (unfiltered) feed returns only events that carry a source URL. watchlist (OpenSanctions PEP listings) is opt-in via kind=watchlist and is currently uncited (source-url NULL for every row).
| Name | Required | Description | Default |
|---|---|---|---|
| kind | No | ||
| limit | No | ||
| since | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, so the safety profile is covered. The description adds valuable behavioral context beyond that: the default feed only returns events with a source URL, and watchlist events are uncited (source-url NULL). This explains filtering behavior and data provenance, which is not present in the annotations. No contradictions found.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is compact, with the core purpose front-loaded in the first sentence. It covers essential points (default filter, watchlist caveat, streaming usage) without excess. The structure is logical: purpose → usage hint → behavioral nuance. No redundant or filler content.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool is a feed with 3 parameters and no output schema. The description explains the feed's content (ownership-change and filed-accounts events), default filtering behavior, watchlist specifics, and the streaming mechanism. While it does not describe return fields or pagination details explicitly, the 'since' usage implies a cursor pattern. Given the tool's moderate complexity and the presence of helpful annotations, the description is fairly complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It explains 'since' (stream new events) and 'kind' (opt-in watchlist, implying default kinds ownership/filing), but does not explain the 'limit' parameter at all. The description adds meaning for two of three parameters, but the missing 'limit' documentation leaves a gap. Given the low coverage, this is partial compensation.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'The corpus activity feed — recent ownership-change and filed-accounts events, cited.' This provides a specific verb (feed) and resource (corpus activity), and the mention of 'cited' adds a distinguishing feature. However, it does not explicitly contrast with sibling tools like search or entity lookups, so it lacks explicit sibling differentiation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description gives usage instructions for the 'since' parameter ('Pass since=<next_since> to stream only new events') and notes that 'watchlist' is opt-in via 'kind=watchlist'. This implies when to use the tool (to stream new events or request watchlist data) but does not provide explicit guidance on when NOT to use it or how it compares to alternatives like search_entities or get_entity. The guidance is contextual but not comprehensive.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have distinct purposes (search, lookup, get details, graph traversal, coverage maps, payment). Some overlap exists between lookup_company and lookup_by_identifier (both resolve identifiers, though one is UK-specific), and between get_entity vs get_company_details, but descriptions clarify distinctions. find_similar and semantic_search both find similar entities but different mechanisms.
Uses a mix of verb_noun and noun_phrases: buy_dossier, claim_dossier, check_offshore_exposure, find_similar, get_pricing, get_pulse, graph_neighbourhood, list_asset_coverage, lookup_by_identifier, search_entities, trace_ownership_path. While most use underscores and verbs like get/lookup/search/list, there are inconsistencies like 'graph_neighbourhood' (noun start) vs 'trace_ownership_path' (verb start). Overall readable but not perfectly consistent.
23 tools is at the high end but justifiable for a comprehensive intelligence server covering search, lookup, graph, coverage, payments, and entity details. 'buy_dossier' is considered one of the tools and adds bulk but not excessive. It is appropriate for the scope.
Covers a wide range: search (search_entities, search_companies, semantic_search), identifier lookup (lookup_by_identifier, lookup_company), entity details (get_entity, get_company_details, get_financials), graph (graph_neighbourhood, graph_path, trace_ownership_path), coverage maps (list_jurisdictions, list_asset_coverage), sanctions (get_sanctions, check_offshore_exposure), plus payment flow (get_pricing, buy_dossier, claim_dossier, get_payment_link). Missing perhaps a 'get_relationships' separate tool but get_entity includes direct relationships. Also no explicit 'list_entities' but search covers it. Minor gaps like no bulk operations besides resolve.
