structureResearch-structure (lab/department) lookup.
| Name | Required | Description | Default |
|---|---|---|---|
| rows | No | 1-1000 (default 25). | |
| query | Yes |
| Name | Required | Description | Default |
|---|---|---|---|
| response | No | Structure search results | |
| responseHeader | No | Response metadata |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Input schema / examplesAdded value: +[
+ {
+ "query": "CNRS"
+ },
+ {
+ "query": "Laboratoire d'Informatique",
+ "rows": 100
+ }
+]Output schema / (root)Previous value: -nullNew value: +{
+ "description": "Research structure reference lookup response from HAL API",
+ "properties": {
+ "response": {
+ "description": "Structure search results",
+ "properties": {
+ "docs": {
+ "description": "Array of matching structure records",
+ "items": {
+ "description": "Structure reference object",
+ "type": "object"
+ },
+ "type": "array"
+ },
+ "numFound": {
+ "description": "Total number of matching structures",
+ "type": "integer"
+ },
+ "start": {
+ "description": "Starting offset of results",
+ "type": "integer"
+ }
+ },
+ "type": "object"
+ },
+ "responseHeader": {
+ "description": "Response metadata",
+ "properties": {
+ "QTime": {
+ "description": "Query execution time in ms",
+ "type": "integer"
+ },
+ "params": {
+ "description": "Query parameters echoed back",
+ "type": "object"
+ },
+ "status": {
+ "description": "HTTP status code",
+ "type": "integer"
+ }
+ },
+ "type": "object"
+ }
+ },
+ "type": "object"
+}Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, destructiveHint=false, and idempotentHint=true, so the safe read-only nature is known. However, the description adds no additional behavioral context—such as whether results are paginated, how to interpret the output, or any rate limiting. It simply restates the tool's function without enriching the annotation-provided safety profile.
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 a single, concise sentence, which is appropriate for a simple tool. It wastes no words and is easy to parse. However, it is slightly under-specified, bordering on terse, but the structure itself is perfectly fine.
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?
Given the output schema exists and the tool is simple, this could be complete with just the basic info. But the description does not explain how to interact with the tool effectively—no mention of query format, result count defaults, or whether the output is a list or single entity. The annotations cover safety but not usage, and the description leaves out practical details.
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 50%: the 'rows' parameter has a description, but 'query' is only typed as string with examples. The tool description adds no parameter-level explanation beyond the schema, and the examples don't clarify query semantics beyond 'CNRS' or 'Laboratoire d'Informatique'—the description fails to compensate for the low coverage.
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 performs a lookup for research structures (labs/departments). The verb 'lookup' and resource 'research-structure' are specific, and the parenthetical clarifies the domain. However, it doesn't differentiate from sibling lookup tools like entity_profile or search, lacking explicit distinction.
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?
No guidance is provided on when to use this tool versus alternatives. The description only implies it's for research-structure lookup, but there are no when-to-use or when-not-to-use instructions, and no mention of alternative tools. This leaves the agent to guess based on the tool name alone.
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.
Several tools occupy nearly the same role: ask_pipeworx, ask_pipeworx_beta, and ask_pipeworx_grounded are deliberately near-duplicates, while deep_research, validate_claim, bet_research, and the polymarket_* family all route factual questions to overlapping data pipelines. Generic single-word tools like get, search, structure, and author add further ambiguity, making it hard for an agent to confidently pick the right tool.
Names are consistently lowercase snake_case, but they mix verb_noun tools (list_subscriptions, resolve_entity, validate_claim) with bare nouns (author, get, search, structure) and domain-prefixed families (ask_pipeworx, polymarket_*, pipeworx_*). The conventions are readable but not predictable enough to infer behavior from the name alone.
35 tools is heavy for a single server, especially when many are meta-routers or near-variants of each other. The broad data-research scope justifies some breadth, but the surface feels padded with overlapping research and prediction-market tools rather than a tight, well-scoped set.
For its apparent purpose — authoritative data lookup, verification, research, entity profiling, prediction-market analysis, and monitoring — the surface is largely complete: retrieval, grounded answers, deep research, comparison, change tracking, subscriptions, and memory are all covered. Minor gaps exist, such as no direct tool for managing alert delivery or for some of the vague HAL-style operations, but agents can work around these via the router tools.