An MCP server that provides timer management tools integrated with Resonate for background job execution. Allows setting timers with names and durations, and checking timer status asynchronously.
The server provides two tools with clear verb-noun naming (set_timer, get_timer_status) and basic descriptions. However, the implementation lacks critical quality patterns: no input parameter descriptions in schemas, no documented output schemas, minimal parameter validation guidance, and no error handling or recovery information. The descriptions are present but generic (63-65 chars) and do not explain WHEN to use the tool, WHAT it returns, or HOW to handle failures. Parameter schemas exist but descriptions are minimal ('The name of the timer', 'The duration of the timer in seconds'), these fail the 50-200 char LLM-optimization baseline and do not specify constraints (e.g., must seconds be positive? What is the max duration?). Tool composition is sound (two focused tools), but overall definition quality is below production grade.
Get the status of a timer by its name.
Set a timer with the given name and duration in seconds.
Parameter descriptions are too brief and lack constraint information. 'The name of the timer' (20 chars) and 'The duration of the timer in seconds' (37 chars) do not specify format, range, or valid values. LLMs cannot infer constraints from these descriptions.
No documented output schema. Tools return dicts with 'promise_id' and 'status' fields respectively, but LLMs have no indication of what fields to expect or their types. This violates the requirement that LLMs know what fields to extract for downstream tool calls.
No error handling or recovery guidance. If a timer cannot be created (invalid name, negative seconds, server error), the tool returns nothing and the LLM has no instructions for what to do next. Missing actionable error messages.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 59 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 45 | - | v1 |
Tool descriptions lack context for LLM selection. They do not explain WHEN to use set_timer vs get_timer_status, what prerequisites exist (must timer be set first?), or whether operations are idempotent. Descriptions should answer: What does it do? When to use it? What does it return?
No input validation constraints. The 'seconds' parameter in set_timer has no specified minimum/maximum. Can an LLM pass 0, -1, or 999999? This invites invalid calls and wastes tokens on retries.
No pagination or result limits documented. While these tools do not return lists, the pattern is relevant if the server is extended. The rubric requires explicit statement of limits in descriptions.