Last N hours of normalized points [{ ts, mg_dl, trend }] from the current graph window (max ~12h). Read-only CGM data — not for medical decisions without clinician review.
Bulk support: accepts patient_ids for batched execution.
| Name | Required | Description | Default |
|---|---|---|---|
| hours | No | ||
| account | No | ||
| patient_id | No | ||
| patient_ids | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate readOnlyHint=true and destructiveHint=false, but the description adds value by disclosing the medical disclaimer and bulk support. It also clarifies data source is 'current graph window (max ~12h)', which is beyond annotations.
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 concise (two sentences) and front-loaded. However, the second sentence about bulk support could be integrated more smoothly. Still efficient and readable.
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 return structure is described, which is good given no output schema. However, parameter documentation is incomplete, and the relationship between 'hours' and 'max ~12h' is unclear. The medical disclaimer fills some context but not all.
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 parameter descriptions have 0% coverage. The description mentions 'hours' and 'patient_ids' but does not explain 'account' or 'patient_id', nor does it specify valid ranges, defaults, or formats. Users must infer from context.
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 returns 'Last N hours of normalized points' with a specified data structure [{ ts, mg_dl, trend }], distinguishing it from siblings like librelink_get_current_glucose (single reading) and librelink_get_glucose_graph (different representation). The resource and action are specific.
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?
Usage context is implied ('current graph window') but no explicit guidance on when to choose this tool over siblings like librelink_get_glucose_graph or librelink_get_glucose_logbook. No 'when not to use' or alternative suggestions.
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 are clearly distinct by function, but librelink_list_accounts and librelink_list_connections are exact aliases, causing confusion. Some glucose tools have overlapping outputs (e.g., librelink_get_current_glucose vs librelink_get_latest_reading), though descriptions help differentiate them.
CGM data tools consistently use the librelink_ prefix, but authentication, connect, marketplace, report_bug, show_version, and toolkit_info lack the prefix, mixing conventions. This inconsistency makes the tool set feel disjointed.
With 16 tools, the set is slightly larger than needed, but it covers the main CGM functionalities plus platform utilities. The number is still reasonable and manageable, with each tool serving a specific purpose.
The CGM data tools provide comprehensive read-only access (current, graph, stats, logbook, etc.), but there are no write operations (e.g., setting targets) or historical data beyond ~12-14 days. The inclusion of marketplace and configuration tools feels out of scope but does not leave critical gaps for the stated domain.