Find and run trustworthy calculators with clear methods, provenance, and sources.
Scored across 3 tools
Each tool has a clearly distinct purpose: schema retrieval, execution, and catalog search. Descriptions explicitly reference each other (e.g., 'inspect get_calculator_schema'), reinforcing correct selection.
All tool names follow a consistent verb_noun pattern: get_calculator_schema, run_calculator, search_calculators. The pattern is predictable and readable.
Three tools cover the core workflow (discover, inspect, execute). While the count is on the low side, each tool is substantial and the scope is well-defined, leaving room for future growth.
The surface covers discovery, schema inspection, and execution, which are the essential operations. However, lifecycle operations like batching multiple runs or saving results are absent, which could limit advanced use cases.
Get the input and output schema for a calculator, generated from its reviewed contract: parameter types, units, allowed dropdown values and definitional bounds, plus outputs keyed by named range (including the status output). Table inputs carry a 'grid' block giving the exact number of rows to send, the column layout, which columns are editable, and how to shape a row; a table whose 'grid' block says writable=false cannot be supplied at all. Schema 0.18 exposes required and omitted declarations plus version-bound visibility tables; older versions require every input while awaiting metadata migration. Optionally resolves one immutable platform version. Returns trust metadata for the resolved version.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | The calculator slug, e.g. 'steel-column-capacity-calculator'. | |
| platform_version | No | Optional positive platform version number. Omit for the current published version. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden, and it delivers: it discloses the writable=false table constraint, the pre-0.18 'every input required' behavior versus 0.18's required/omitted declarations, and that it returns trust metadata for the resolved version. It doesn't cover auth requirements or rate limits, keeping it below a 5.
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?
Front-loaded with purpose, then details about the returned schema structure and version behavior. Dense but every clause adds information; the middle sentence about 'grid' blocks is long but earns its place for a schema-inspection tool.
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?
No output schema exists, so the description must characterize returns — and it does, describing parameter types, units, dropdown values, bounds, named-range outputs, the status output, and grid blocks. It also covers the version-visibility caveat. Minor gaps remain (no example call, no errors), but the return-shape coverage is notably thorough.
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 100% — both slug and platform_version are fully documented in-schema, including the 'omit for current published version' default and the 'must be positive' constraint. The description adds only the 'one immutable platform version' framing, so baseline 3 applies.
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+resource ('Get the input and output schema for a calculator') and names the source of truth ('generated from its reviewed contract'). This unambiguously distinguishes it from run_calculator (execute) and search_calculators (discover).
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 implies this is the inspection step before invocation, and mentions the optional platform_version resolution, but never explicitly states when to call this versus run_calculator or how it fits a workflow. Usage is inferable rather than spelled out.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Run a calculator with the given inputs. Inputs are validated against the resolved version's reviewed contract and computed by SpreadsheetWeb through the same proxy as the web channel. An optional platform_version pins both validation and computation to that immutable version. Returns outputs keyed by named range (including the status output) and trust metadata with a citation URL, resolvedInputs, and warnings for omitted inputs. Missing or blank visible required inputs are refused before computation; inspect get_calculator_schema and the complete error detail. For schema 0.18, optional omissions are sent explicitly as declared blank or standing default and disclosed with their values and sources. Older pinned contracts require all inputs during migration. Visibility follows version-bound dropdown truth tables; a missing controller makes its dependents indeterminate until it is supplied. A hidden controller is ignored only when all of its option branches have the same remaining visibility subtree; otherwise its dependents stay hidden. Hidden omissions use explicit blank/default values; supplied hidden inputs are retained with a not-applicable warning. Always read Model_Status and status outputs; CHECK and STOP are load-bearing verdicts about the result's validity. Results are informational and not professional advice; the applicable notice is returned in the trust block.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | The calculator slug. | |
| inputs | Yes | Input name/value pairs matching the calculator schema (from get_calculator_schema). A table input takes a list of rows matching its 'grid' block exactly; a table of the wrong extent or shape is refused rather than partially applied. | |
| platform_version | No | Optional positive platform version number. Omit for the current published version. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden and does so richly: refusal of missing/blank required inputs, explicit blank/default handling for omissions, visibility truth-table and hidden-controller rules, retention of supplied hidden inputs with a not-applicable warning, and trust metadata with citation URL. It also flags that Model_Status/CHECK/STOP are load-bearing verdicts, which is essential behavioral context an agent could not infer.
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?
Front-loaded with purpose and return shape, but the body is a dense wall of edge-case prose (truth tables, hidden controllers, migration behavior) that is hard to scan. The critical 'read Model_Status' guidance is buried mid-paragraph rather than surfaced.
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?
No output schema exists, so the description must cover returns — and it does, specifying outputs keyed by named range, the status output, and a trust block with citation URL, resolvedInputs, and warnings. For a complex nested-input tool this is complete enough to invoke correctly.
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 100%, so baseline is 3, but the description adds real meaning: platform_version 'pins both validation and computation to that immutable version', and inputs must match the schema returned by get_calculator_schema, including exact table grid extents. This goes beyond the schema's own field descriptions.
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?
Opens with a specific verb+resource ('Run a calculator with the given inputs') and distinguishes itself from siblings by directing the agent to get_calculator_schema for input discovery. An agent can tell this is the execution tool versus the schema/search tools.
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?
Gives clear context: inspect get_calculator_schema first, platform_version pins an immutable version, missing required inputs are refused. It stops short of explicitly naming when NOT to use it or contrasting with siblings beyond the schema-lookup pointer, but the operational context is strong.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search the LogicCommons calculator catalog by keyword and optional category. Returns matching published calculators with slug, name, summary, category, a citation URL, and trust metadata.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | Free-text query matched against calculator name, summary and category. Empty returns the catalog. | |
| category | No | Optional category slug to filter by, e.g. 'structural-civil'. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full behavioral burden; it discloses that only published calculators are returned and that results carry trust metadata and a citation URL, which is useful. However, it says nothing about pagination, result limits, or auth requirements for a catalog search.
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?
Two sentences, zero filler, with the core action and scope front-loaded before the return-value summary. Every clause earns its place.
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 simple two-parameter read search with no output schema, the description usefully enumerates the returned fields (slug, name, summary, category, citation URL, trust metadata). Only pagination/limit behavior and explicit sibling routing are missing.
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 100%, so both parameters are already fully documented, including the empty-query behavior and the category slug example. The description only restates 'keyword and optional category' and adds no syntax or matching semantics beyond the schema.
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 names a specific verb (search) and resource (LogicCommons calculator catalog) and scopes it by keyword and category. It is clearly distinguishable from a schema-fetch or execution tool, though it never names its siblings get_calculator_schema/run_calculator to draw the boundary explicitly.
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?
There is no when-to-use or when-not-to-use guidance and no mention of the alternative sibling tools. The only usage hint ('Empty returns the catalog') lives in the schema, not the description.
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_calculator_schemarun_calculatorsearch_calculators140+ calculators and data tools — finance, health, science, global comparisons and more.
Free calculators as MCP tools: finance, taxes, health, units, dates. Search, fetch & compute.
AI-callable calculators and engineering models with real formulas. No hallucinated math.
Rebuilds the scores real systems run on you — credit, actuarial, lending — in the open, cited.
Add one secure layer between your agents and this server.
