Données eDNA marines Guadeloupe MNHN 2021-2022 : 300+ espèces poissons, 21 cétacées, protocole eDNA. Prix : 0.100 USDC.
| Name | Required | Description | Default |
|---|---|---|---|
| payment_tx_hash | No | Hash tx USDC Base. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
While readOnlyHint already indicates a safe read operation, the description adds important behavioral context by disclosing the data source, species counts, protocol, and a price of 0.100 USDC, implying a paid transaction. This goes beyond the annotation and helps set expectations about cost and data provenance.
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, compact sentence packed with useful specifics: geographic area, institution, year range, species counts, protocol, and price. No filler or redundancy.
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 data-retrieval tool with one optional parameter, the description provides good context about the data content and pricing. However, it lacks details on how to actually obtain the data (e.g., is payment_tx_hash required to trigger data delivery?) and does not mention return format or authentication, leaving some ambiguity for an agent.
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% for the single optional parameter payment_tx_hash, which already explains it is a USDC Base transaction hash. The description's mention of price (0.100 USDC) indirectly relates to the parameter but does not clarify whether the hash is required, optional, or how it is used.
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 identifies the tool as providing marine eDNA data for Guadeloupe from MNHN 2021-2022, with specific details on species counts and protocol. While no explicit verb like 'retrieve' is used, the content strongly implies a data access tool and distinguishes it from sibling tools via the eDNA topic.
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 given on when to use this tool versus alternatives such as caribbean_get_biodiversite or other data access tools. The mention of price is transactional but does not help an agent decide between this and related tools.
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 clearly distinct domains (solar, biodiversity, water, etc.), but the pair caribbean_get_marches_publics and caribbean_get_marches_publics_dom overlap significantly, differing only in geographic scope. The sub-group of 'caye' tools (marche, vanille, vetiver) are also thematically similar, though descriptions clarify their specific commodities. Overall, an agent could generally select correctly, but a few boundary cases exist.
The tools overwhelmingly follow a caribbean_get_<subject> convention, with only two exceptions: caribbean_analyse_solaire (uses 'analyse' instead of 'get') and caribbean_list_marketplace (uses 'list' instead of 'get'). This makes the pattern predictable and readable, with minor deviations that are still intuitive. The subject names are consistently descriptive, though some use abbreviations like 'dom' and 'oa' that might require context.
24 tools is on the heavy side for a single server, particularly given that many tools are single-purpose data lookups (e.g., biodiversity, eDNA, tourism). The count is justifiable for a wide-ranging Caribbean data API, but it risks overwhelming an agent with a large surface. It is not excessive enough to be absurd, but it's in the 'heavy' zone.
The server covers multiple domains comprehensively, such as solar energy (tariffs, irradiation, connection, tax credit, financing) and public contracts (Guadeloupe and all DOM). The marketplace meta-tools (llms_txt, list_marketplace, get_provider) ensure discoverability, but there is no general search or aggregation tool. Some areas like agriculture or real estate beyond what's present could be considered gaps, but the core coverage is solid.