Skip to main content
Glama
Sign In

HypePaper

by app.hypepaper
OverviewDiscussionsAdmin

sota_alarm

Standing alarm on state of the art: watch a topic or a benchmark lane, then poll for the record changes you have not seen yet. action='watch' adds a watch (topic, or dataset with an optional metric), action='unwatch' removes one, action='poll' (the default) returns new events across every watch — the previous holder, the new leader, the delta, and how long the old record stood. A poll marks what it returned as seen and moves the cursor, so the next call is a true diff; pass peek=true to look without consuming. Use it to keep a project current instead of re-reading a leaderboard: one call a day tells you if anything moved. [Free · 10 credits]

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "sota_alarm"
  }
}

Input Schema

NameRequiredDescriptionDefault
peekNoReturn events without marking them seen
limitNoEvents to return
topicNoTopic name or UUID to watch (its whole subtree)
actionNoDefault 'poll'. Every action also returns the new events.poll
metricNoNarrow a dataset watch to one metric, e.g. 'Accuracy'
datasetNoDataset name or key to watch, e.g. 'ImageNet'
watch_idNoThe watch to remove (from 'watches')
conditionNoExact experimental protocol key; omitted watches each protocol separately

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A3.9/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does well: it discloses that poll is stateful (marks returned events seen, moves the cursor), that peek is the non-consuming read, that every action also returns the new events, and it states cost ('Free · 10 credits'). It does not cover pagination limits or error/permission behavior, which the schema only partly implies via limit.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A dense single paragraph that front-loads the core concept and the three actions before the polling semantics, then closes with usage guidance and cost. Nearly every sentence earns its place; only the pricing tag is tangential but still decision-relevant.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For an 8-parameter, stateful tool with no annotations and no output schema, the description covers action semantics, cursor consumption behavior, and what a poll returns. Left unexplained are pagination behavior for the 50-event cap and how multiple watches interact with limit, but the essentials for correct invocation are present.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3. The description adds genuine relational meaning not in the schema — that a watch is either a topic or a dataset with an optional metric — but it does not explain watch_id, condition, or limit behavior beyond what the schema already says.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a concrete verb set (watch / unwatch / poll) over a concrete resource (topics and dataset benchmark lanes) and states the outcome returned: previous holder, new leader, delta, and record duration. It is clear what the tool does, though it does not name a specific sibling (e.g. sota_changelog, sota_check) to contrast against.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives explicit operating context — 'keep a project current instead of re-reading a leaderboard: one call a day' — and explains the poll-vs-peek choice and the default action. It lacks an explicit when-not or a named alternative tool, so it stops short of a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All HypePaper toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources