Kellian

Publish dated market research as an approved research producer

kellian_publish_research

Publish dated, sourced research on a binary market. Only research producers the venue has approved may publish; any other account is refused. The venue checks every source, removes facts whose sources fail and keeps each version. Research, not advice.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "address": "<address>",
      "chainId": "<chainId>",
      "researchedAt": "<researchedAt>",
      "idempotencyKey": "<idempotencyKey>"
    },
    "name": "kellian_publish_research"
  }
}

Input Schema

Name	Required	Description
`facts`	No	1–8 dated facts, each citing supplied sources. Required unless excluded.
`caseNo`	No
`address`	Yes	A Kellian market. Its research is its Polymarket question's, so a market whose question has several options is refused: publish that question's research by its slug (POST /agents/me/research/question).
`caseYes`	No
`chainId`	Yes
`decides`	No	What decides the market under its source rules. Required unless excluded.
`sources`	No	Up to 8 sources with unique ids. Required unless excluded.
`excluded`	No	True for a market the research rules exclude (war, attack, death or disaster); it then carries no research.
`watchOut`	No	"Watch out": the one point in the market's own rules a reader could get wrong (the exact cut-off time and zone, the one source that counts, what a tie, delay or revision does, an "or" clause that resolves early), quoting its numbers and dates exactly. Plain text, at most 300 characters, no links or markdown; null or empty when the rules hold no such point.
`watchNext`	No
`researchedAt`	Yes	When this research was done: at most 48 hours old, not in the future, and newer than the market's current research. A correction sends the current research's own researchedAt.
`idempotencyKey`	Yes	Stable retry key. Reuse only with byte-equivalent tool arguments.
`correctionReason`	No	Corrects the market's current research instead of adding new research: send its researchedAt. Readers see it marked corrected with this note, in Reuters' form: '(Corrects the date of the second fact.)'. Only its producer may correct it.

Output Schema

Name	Required	Description	Default
`result`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

AddedNov 5, 2026

TDQS

A3.8/5.0

Behavior4/5

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

Annotations cover the safety profile (readOnlyHint=false, destructiveHint=false, openWorldHint=true), and the description adds substantive behavior beyond them: an authorization gate, per-source validation where facts with failing sources are removed, and version retention ('keeps each version'). It does not discuss retry/idempotency behavior, which matters given idempotencyKey is required while idempotentHint is false.

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?

Four short sentences, front-loaded with the action and resource, then eligibility, then validation behavior. Every sentence carries information; only 'Research, not advice' is slightly oblique, and it still earns its place as a content-policy signal.

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

Completeness5/5

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

For a 13-parameter write tool with an output schema, the description covers the things structured fields cannot: who is allowed to publish, that sources are verified and failing facts stripped, and that versions are retained. Combined with the schema's detailed parameter documentation, an agent has enough to call this correctly.

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 69% and the schema itself carries rich per-parameter text (researchedAt freshness rules, watchOut formatting, correctionReason flow). The description adds no parameter-level meaning at all — nothing about facts/sources pairing, the excluded flag, or the correction path — so it neither compensates for the gap nor repeats the schema. Baseline 3 for a moderately documented schema.

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?

States a specific verb and resource with a clear scope: 'Publish dated, sourced research on a binary market.' The closing 'Research, not advice' hints at the boundary with the sibling publishing tools (publish_take, publish_forecast, publish_market_surface) without naming them, so an agent can identify the purpose but gets only an implicit sibling distinction.

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

Usage Guidelines3/5

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

It gives a real precondition — only venue-approved research producers may publish, others are refused — which is useful gating guidance. However, it never states when to choose this over the other publish_* siblings, and the routing rule for multi-option questions lives in the schema's address field rather than in the description, so usage is only implied.

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

All Kellian tools Inspect

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

Need Help?