> For the complete documentation index, see [llms.txt](https://docs.biobox.io/guide/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.biobox.io/guide/how-to/ask-biobox-with-claude/tools/get-reasoning-model.md).

# get\_reasoning\_model — Get reasoning model config

Get reasoning model config

Get reasoning model config

## What it does

Open ONE reasoning model (either topology) and return its scoring CONFIG — its evidence lines (each with a weight, the query that defines it, and scoring node), the evidence categories they span, the anchor/returns concepts, and the model's workflow status + log links.

**When to use:** the user asks "what evidence does model X use / what is it scoring on, and how is each weighted?"; you want to understand HOW a model scores (its evidence lines + weights + queries) before reading its reports; you have a model's uuid from list\_reasoning\_models and need its detailed configuration; the user asks whether a model is up to date / has finished running, or wants its workflow logs.

**When NOT to use:** listing ALL the org's models / finding which model fits (use list\_reasoning\_models); a model's reports / report inventory (use list\_reports); the actual ranked scores a report produced (use get\_report\_scores); why a SPECIFIC entity (e.g. one gene) scored the way it did (use get\_evidence).

**Parameters:**

* `modelId` — the model's uuid (from list\_reasoning\_models `models[].uuid`) — identifies which model's config to open. Works for both bipartite and monopartite/profile models.

**Returns:** The model config, flat at the root: displayName, description, folder (org grouping, or null), topology (bipartite | monopartite), anchorConcept (the class an answer is anchored on) and returnsEntityType (the class the model ranks — null for monopartite/profile) — each a { dbLabel, displayName } ref (dbLabel is the graph label for Cypher, displayName the human name), workflowStatus (Succeeded / Out of date / Running / Failed), graphUpToDate (boolean), hasArchivedReports (true if the model has archived scoring reports — call list\_reports with includeArchived to see them), workflowLogs ({ stdoutUri, stderrUri }), evidenceLineCount, categories (distinct evidence categories, sorted), and evidenceLine — the full list of scoring evidence lines, each { uuid, name, description, category, weight, bxql, cypher, scoringNode }. The per-line `weight` is how much that line contributes to the score; `bxql` is the query the author wrote and `cypher` is the graph path compiled from it; `scoringNode` is the node it scores. When the id resolves to no model, the tool returns a 404 Not Found error (isError) naming the missing id.

## Parameters

| Parameter | Type   | Required | Description                                                                                                                                                         |
| --------- | ------ | -------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `modelId` | string | yes      | the model's uuid (from list\_reasoning\_models `models[].uuid`) — identifies which model's config to open. Works for both bipartite and monopartite/profile models. |

## Response

```ts
{
  displayName: string;
  description: string | null;
  folder: string | null;
  topology: "bipartite" | "monopartite";
  anchorConcept: {
    dbLabel: string | null;
    displayName: string | null;
  } | null;
  returnsEntityType: {
    dbLabel: string | null;
    displayName: string | null;
  } | null;
  workflowStatus: "Failed" | "Out of date" | "Running" | "Succeeded";
  graphUpToDate: boolean;
  hasArchivedReports: boolean;
  workflowLogs: {
    stdoutUri: string | null;
    stderrUri: string | null;
  };
  evidenceLineCount: number;
  categories: string[];
  evidenceLine: Array<{
    uuid: string;
    name: string;
    description: string | null;
    category: string;
    weight: number;
    bxql: string | null;
    cypher: string | null;
    scoringNode: string | null;
  }>;
}
```

## Example

```json
// get_reasoning_model({"modelId":"9GFXylq5eiFxkbc0XugAR"})
{
  "modelId": "9GFXylq5eiFxkbc0XugAR"
}
```

Response (arrays truncated):

```json
{
  "displayName": "Indication Prioritization",
  "description": "Identifying indications for lead asset",
  "folder": null,
  "topology": "monopartite",
  "anchorConcept": {
    "dbLabel": "Disease",
    "displayName": "Disease"
  },
  "returnsEntityType": null,
  "workflowStatus": "Succeeded",
  "graphUpToDate": true,
  "hasArchivedReports": false,
  "workflowLogs": {
    "stdoutUri": "profileKernelWorkflow/9GFXylq5eiFxkbc0XugAR/hkJELM-4vfq0BrLWhNnI6/wfstout.txt",
    "stderrUri": "profileKernelWorkflow/9GFXylq5eiFxkbc0XugAR/hkJELM-4vfq0BrLWhNnI6/wfsterr.txt"
  },
  "evidenceLineCount": 4,
  "categories": [
    "Biomarkers",
    "Clinical Trial",
    "Drugs",
    "… 1 more"
  ],
  "evidenceLine": [
    {
      "uuid": "sW9tmrPbzY1WFpvVFq8iL",
      "name": "Drug approved for the disease",
      "description": "Drug is approved for the disease",
      "category": "Drugs",
      "weight": 0.2,
      "bxql": "(scoringAndUniqueNode:Drug) -> is approved for -> (anchorConcept:Disease)",
      "cypher": "(scoringAndUniqueNode:`Drug`)-[_r:`is approved for`]->(anchorConcept:`Disease`)",
      "scoringNode": "Drug"
    },
    {
      "uuid": "Dbi96cYNlgnzNJbBl_9Yg",
      "name": "Disease is studied in a clinical trial",
      "description": "Number of clinical trials currently studying the disease",
      "category": "Clinical Trial",
      "weight": 0.8,
      "bxql": "(anchorConcept:Disease) <- studies condition <- (scoringAndUniqueNode:ClinicalTrial)",
      "cypher": "(anchorConcept:`Disease`)<-[_r:`studies condition`]-(scoringAndUniqueNode:`ClinicalTrial`)",
      "scoringNode": "ClinicalTrial"
    },
    {
      "uuid": "r097I5_Vj4koqEKRfAq8t",
      "name": "Biomarker is available",
      "description": "A gene is a marker for the disease",
      "category": "Biomarkers",
      "weight": 0.5,
      "bxql": "(scoringAndUniqueNode:Gene) -> is marker for -> (anchorConcept:Disease)",
      "cypher": "(scoringAndUniqueNode:`Gene`)-[_r:`is marker for`]->(anchorConcept:`Disease`)",
      "scoringNode": "Gene"
    },
    "… 1 more"
  ]
}
```