Irrational is an MCP server published by io.github.urbanmorph, hosted at irrational.pages.dev. When last checked on 2026-11-21 it was working: it answers and lists its tools. It declares 3 tools, including analyze_decision, list_biases, get_bias.
Adversarial behavioural-bias engine: audits your decisions for cognitive biases via your own AI.
1 connections through 5 shared values. Tap any node.
Inferred from the host name and headers: what a server runs on, not who runs it.
Adversarially audit a decision for cognitive biases. Returns a directive YOUR model executes to produce the composed audit (verdict-first). Provide reasoning, not just the conclusion.
| Parameter | Type | Description |
|---|---|---|
| mode | string | forward = a decision you are about to make; retrospective = reviewing a past decision/outcome. |
| judgment * | string | The decision/judgment in one line. |
| language | string | Optional. Natural language for the audit prose (e.g. "Tamil", "Spanish"). Bias ids stay canonical English so the result is still parseable. Defaults to English. |
| reasoning | string | How you arrived at it (required to audit). |
| structured | boolean |
None yet. Changes appear after a server's second daily check.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
| Optional. Default false → the audit comes back as readable prose. Set true to get a machine-parseable JSON object (bias ids/keys in English) for pipelines that store or compare audits. |
{
"name": "analyze_decision",
"description": "Adversarially audit a decision for cognitive biases. Returns a directive YOUR model executes to produce the composed audit (verdict-first). Provide reasoning, not just the conclusion.",
"inputSchema": {
"type": "object",
"required": [
"judgment"
],
"properties": {
"mode": {
"enum": [
"forward",
"retrospective"
],
"type": "string",
"description": "forward = a decision you are about to make; retrospective = reviewing a past decision/outcome."
},
"judgment": {
"type": "string",
"description": "The decision/judgment in one line."
},
"language": {
"type": "string",
"description": "Optional. Natural language for the audit prose (e.g. \"Tamil\", \"Spanish\"). Bias ids stay canonical English so the result is still parseable. Defaults to English."
},
"reasoning": {
"type": "string",
"description": "How you arrived at it (required to audit)."
},
"structured": {
"type": "boolean",
"description": "Optional. Default false → the audit comes back as readable prose. Set true to get a machine-parseable JSON object (bias ids/keys in English) for pipelines that store or compare audits."
}
}
},
"annotations": {
"title": "Audit a decision for biases",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}List the 22-bias catalogue, optionally filtered by family.
| Parameter | Type | Description |
|---|---|---|
| family | string | Optional. Only the biases in this family; omit for all 22. |
{
"name": "list_biases",
"description": "List the 22-bias catalogue, optionally filtered by family.",
"inputSchema": {
"type": "object",
"properties": {
"family": {
"enum": [
"too-much-information",
"not-enough-meaning",
"need-to-act-fast",
"what-we-remember"
],
"type": "string",
"description": "Optional. Only the biases in this family; omit for all 22."
}
}
},
"annotations": {
"title": "List the bias catalogue",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Get the full entry for one bias by id.
| Parameter | Type | Description |
|---|---|---|
| id * | string | The bias id, as list_biases returns it (e.g. "anchoring"). |
{
"name": "get_bias",
"description": "Get the full entry for one bias by id.",
"inputSchema": {
"type": "object",
"required": [
"id"
],
"properties": {
"id": {
"type": "string",
"description": "The bias id, as list_biases returns it (e.g. \"anchoring\")."
}
}
},
"annotations": {
"title": "Get one bias",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Irrational is an adversarial behavioural-bias auditor. It runs no model of its own: `analyze_decision` returns a DIRECTIVE, an adversarial contract, the bias catalogue and a procedure. You must EXECUTE that directive yourself to produce the audit; do not just print or summarise it. Follow its instructions exactly: assume the user is biased, refuse to validate, argue the other side, and tie every bias to the user's own words. Call analyze_decision with the user's judgment and the reasoning behind it (ask for the reasoning first if only a conclusion is given). Output is readable prose by default; pass structured:true for a parseable JSON object, and language to answer in another language. Use list_biases / get_bias to reference or explain the 22-bias catalogue.
Names, descriptions and schemas were written by the server and are shown as data.