Research on UK accountancy companies from Companies House data: ownership, size, age, filings.
Scored across 1 tool
With only one tool there is no tool-level ambiguity at all; the only selection burden is the dataset parameter, whose three modes are described inline. It loses a point because cramming three distinct datasets behind a single mode parameter is less clear than separate tools would be.
The single name uses clean snake_case with a clear verb_noun shape. There is no pattern to break, but also no pattern to demonstrate, so it can't reach a full 5.
One tool is very thin for a research-data server covering three datasets (ownership/size/age, filing health, and filing calendar). A single multiplexed tool forces agents to parse a mode parameter to discover capabilities.
The surface covers read access to three aggregate datasets, which is a coherent scope. However, it is read-only with aggregate-only figures and no per-company lookup, filtering, or time-range options, leaving notable gaps for typical research workflows.
Return one of the research datasets: "research" (ownership, size, age), "filing-health" (overdue filings) or "accounts-calendar" (filings due by month). Aggregate figures only, from Companies House data.
| Name | Required | Description | Default |
|---|---|---|---|
| dataset | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It does add real behavioral context: results are "aggregate figures only" and sourced from "Companies House data", which sets expectations about granularity and provenance. It omits freshness/caching, auth requirements, and whether the three datasets differ in cost or latency.
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?
A single sentence, front-loaded with the action and resource, then the enumerated options. Every clause carries information; no filler. The nested quotes/colons make it slightly dense to parse but nothing is wasted.
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?
For a one-parameter read tool with no output schema, the description covers purpose, the meaning of the sole parameter, and the granularity/source of the data. The remaining gap is what the returned aggregate figures actually look like, which matters more given there is no output schema.
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 coverage is 0%, so the description must compensate — and it does, mapping each of the three enum values to the content it returns. This is exactly the missing information an agent needs to choose a value. It stops short of describing response shape or sizing per dataset.
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?
States a specific verb ("Return") and resource ("one of the research datasets") and then enumerates the three dataset variants with a parenthetical gloss for each. An agent can tell exactly what it gets for each input value. No siblings exist to differentiate from, so 4 rather than 5.
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 per-dataset glosses ("ownership, size, age", "overdue filings", "filings due by month") implicitly tell an agent which value to pick for which question. However, there is no explicit when-to-use framing, no prerequisites, and no statement of what this tool is not for.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_accountancy_researchUK Companies House due diligence: beneficial owners, directors, ECCTA, accounts. Hosted, no key
UK due diligence — Companies House, Charity Commission, Land Registry, Gazette, HMRC VAT
UK company data: search, officers, owners, shareholders, filed accounts, new companies, KYB checks.
Every UK company plus 35M Companies House accounts since 2008: lookup, search, alerts.
Add one secure layer between your agents and this server.