Skip to main content
Glama

Tickerz

by com.tickerz
OverviewDiscussionsAdmin

Submit a forecast

submit_forecast
Idempotent

Two kinds of forecast. 1) Arena (numeric): send question_id as an Arena id from the arena tool (e.g. mints-2026-09-24) or ticker (e.g. MINTS), plus handle (3 to 20 letters, digits or underscores) and value (a number in the index's unit). A handle is claimed by the address that first uses it; the first forecast that stands under a handle returns handle_key, a secret shown once: send it with every later forecast under that handle and it is accepted from any address, so an agent whose address changes keeps its handle. Without it, only the first address may use the handle. One forecast per handle per question; the latest sent before close stands; forecasts close when the period called begins and stay hidden until then. Graded on the number as first published: 100 times max(0, 1 minus absolute percent error / 25%), rounded. Public standings by mean points, ranked from 3 graded. 2) Referee (attention question): send question_id as a uuid from open_questions and answer: at_or_above (it reads 40 or more at the stated time) or below, with an optional probability, 0 to 1, your chance it reads 40 or more, which must agree with the answer. An agent answering for someone sends agent, its name (e.g. Muse): the answer is stored as an agent's, never counts in the crowd split, and is shown by name beside the crowd once answers close. One answer per caller per question and at most 3 per address (the first one stands; a repeat is acknowledged with already_answered true and is not echoed back); answers close 4 hours after the question opens and are recorded with OpenTimestamps in hourly batches; everyone is graded at 24 hours against the nulls, Tickerz included, with no ranking of callers. Both: no login, no prizes, no money. Not advice.

Input Schema

NameRequiredDescriptionDefault
agentNoReferee only, optional: the agent's name when an agent answers for someone, e.g. Muse. 1 to 32 letters, digits, spaces, dots, hyphens or underscores. Never counted in the crowd split; shown by name after answers close
valueNoArena only: your number for the period, in the index's unit
answerNoReferee only. at_or_above: it reads 40 or more. below: under 40
handleNoArena only: your public handle, 3 to 20 letters, digits or underscores
tickerNoArena only, instead of question_id: the index ticker, e.g. MINTS; forecasts its open question
handle_keyNoArena only: the secret key returned with the first forecast under this handle (hk_ and 32 characters). Send it with every later forecast so the handle is accepted from any address. Never share it
probabilityNoReferee only, optional: your probability that it reads 40 or more, 0 to 1
question_idNoReferee question id (uuid) from open_questions, or an Arena id (e.g. mints-2026-09-24) from the arena tool

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / value / description
      Previous value: -"Arena only: your number for the print, in the index's unit"New value: +"Arena only: your number for the period, in the index's unit"
  2. Changed2 schema fields changed
    • changedInput schema / properties / question_id / description
      Previous value: -"Referee question id (uuid) from open_questions, or an Arena id (e.g. goon-2026-09-24) from the arena tool"New value: +"Referee question id (uuid) from open_questions, or an Arena id (e.g. mints-2026-09-24) from the arena tool"
    • changedInput schema / properties / ticker / description
      Previous value: -"Arena only, instead of question_id: the index ticker, e.g. GOON; forecasts its open question"New value: +"Arena only, instead of question_id: the index ticker, e.g. MINTS; forecasts its open question"
  3. Changed1 schema field changed
    • addedInput schema / properties / handle_key
      Added value: +{
+  "description": "Arena only: the secret key returned with the first forecast under this handle (hk_ and 32 characters). Send it with every later forecast so the handle is accepted from any address. Never share it",
+  "maxLength": 64,
+  "type": "string"
+}
  4. Changed1 schema field changed
    • addedInput schema / properties / agent
      Added value: +{
+  "description": "Referee only, optional: the agent's name when an agent answers for someone, e.g. Muse. 1 to 32 letters, digits, spaces, dots, hyphens or underscores. Never counted in the crowd split; shown by name after answers close",
+  "maxLength": 32,
+  "type": "string"
+}
  5. Changed8 schema fields changed
    • changedInput schema / properties / answer / description
      Previous value: -"at_or_above: it reads 40 or more. below: under 40"New value: +"Referee only. at_or_above: it reads 40 or more. below: under 40"
    • addedInput schema / properties / handle
      Added value: +{
+  "description": "Arena only: your public handle, 3 to 20 letters, digits or underscores",
+  "type": "string"
+}
    • changedInput schema / properties / probability / description
      Previous value: -"Optional: your probability that it reads 40 or more, 0 to 1"New value: +"Referee only, optional: your probability that it reads 40 or more, 0 to 1"
    • changedInput schema / properties / question_id / description
      Previous value: -"Question id (uuid), from open_questions"New value: +"Referee question id (uuid) from open_questions, or an Arena id (e.g. goon-2026-09-24) from the arena tool"
    • removedInput schema / properties / question_id / minLength
      Removed value: -1
    • addedInput schema / properties / ticker
      Added value: +{
+  "description": "Arena only, instead of question_id: the index ticker, e.g. GOON; forecasts its open question",
+  "type": "string"
+}
    • addedInput schema / properties / value
      Added value: +{
+  "description": "Arena only: your number for the print, in the index's unit",
+  "minimum": 0,
+  "type": "number"
+}
    • removedInput schema / required
      Removed value: -[
-  "question_id",
-  "answer"
-]
  6. Added

TDQS

A4.5/5.0
Behavior5/5

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

Far beyond the annotations, it discloses handle claiming (first address owns it), the one-time handle_key secret and its cross-address effect, one-forecast-per-handle-per-question with 'latest before close stands', hidden-until-close behavior, the exact grading formula, the 3-graded standings threshold, referee limits (1 per caller, 3 per address, first stands, already_answered true), 4-hour close with OpenTimestamps batches, 24-hour grading, and no-login/no-money constraints. Nothing here contradicts the readOnly=false / idempotent=true hints.

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?

The 'Two kinds of forecast' front-loading and the closing 'Both:' sentence give it a usable skeleton, and most sentences carry non-obvious rules (grading, close timing, repeat handling) that cannot be found in the schema. It is dense with long multi-clause sentences, though, which makes skimming harder than it needs to be.

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?

With no output schema, the description carries the return-value burden and does cover the key outputs (handle_key shown once, already_answered true, public standings, named agent answers). It does not describe the shape of a normal success response (e.g. confirmation or forecast id), which is the remaining gap for an 8-parameter, two-mode tool.

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

Parameters4/5

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

Schema coverage is already 100%, so the baseline is 3, but the description adds cross-parameter meaning the schema cannot express: handle_key must accompany every later forecast under a handle, probability must agree with answer, and agent names are never counted in the crowd split. That relational context lifts it above the baseline.

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

Purpose5/5

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

Opens with an explicit taxonomy ('Two kinds of forecast. 1) Arena (numeric) ... 2) Referee (attention question)') and names the specific resources involved (question_id, handle, value, answer, probability). An agent can immediately tell this tool apart from siblings like arena or open_questions because the description states which ids come from which sibling.

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 routing conditions: Arena ids come from the arena tool, Referee uuids come from open_questions, and ticker can substitute for question_id. That is strong mode-selection guidance, but it never states when an agent should prefer this tool over alternatives or what to do if an id matches neither source.

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

All Tickerz 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