search_cases

List the cases (saker) handled in a session, optionally filtered by a keyword. Paged: the response has `total` matching cases and returns at most `limit` from `offset`; a `note` says which offset fetches the next page when there is one.

Direct GET

For navigating agents: open the HTML page with parameters in the URL. The result is fetched and displayed without screenshots or form filling.

https://allemannsdata.com/wiki/kilder/stortinget/search_cases/?session_id={session_id}&query={query}&limit=100&offset=0

Replace {session_id}, {query}, {limit}, {offset} with values. Remove unused optional parameters from the URL.

For software and dashboards: Read the operation descriptor. The JSON GET template is https://allemannsdata.com/wiki/api/v1/kilder/stortinget/search_cases?session_id={session_id}&query={query}&limit=100&offset=0.

Parameter	Requirement	JSON type	Description
`session_id`	required	string	A session id from list_sessions(), e.g. "2025-2026".
`query`	optional	union	Optional keyword, matched case-insensitively against each case's title, short title and topic names (upstream has no text search; filtering happens here).
`limit`	optional	integer	Max cases to return (default 100).
`offset`	optional	integer	Skip this many matching cases (for paging, default 0).

Generated JavaScript, Python and response structure

Use JSON when the user requests code, machine-readable data, value extraction or further computation. Use the HTML GET page to identify the correct source and verify parameters.

Example for generated JavaScript and Python

https://allemannsdata.com/wiki/api/v1/kilder/stortinget/search_cases?session_id={session_id}&query={query}&limit=100&offset=0

Replace URL placeholders and remove unused optional parameters.

JavaScript in a dashboard

const url = "https://allemannsdata.com/wiki/api/v1/kilder/stortinget/search_cases?session_id={session_id}&query={query}&limit=100&offset=0";
const response = await fetch(url);
if (!response.ok) {
  const failure = await response.json();
  throw new Error(`${failure.error.code}: ${failure.error.message} ${failure.suggestion}`);
}
const result = await response.json();
const data = result.data;

Python without additional packages

import json
from urllib.request import urlopen

url = "https://allemannsdata.com/wiki/api/v1/kilder/stortinget/search_cases?session_id={session_id}&query={query}&limit=100&offset=0"
with urlopen(url) as response:
    result = json.load(response)
data = result["data"]

Response structure

{
  "source": "stortinget",
  "operation": "search_cases",
  "parameters": {},
  "data": {},
  "entities": [],
  "_meta": {"retrieved_at": "...", "sources": [], "cache": {}},
  "_links": {"self": "...", "source": "...", "wiki": "..."}
}

Form for manual use

The form is a fallback for people and browser agents that cannot open the GET URL directly.

Continue exploring

How an AI agent should use this page

Ground rule: Navigate with HTML GET links to keep explanations and relationships. Use the page's JSON alternative only when concrete values must be extracted, compared or computed.

How to use the search_cases lookup

Replace the placeholders in the HTML GET template; remove optional parameters you do not need.
Check types, units, limits and ID formats in the parameter table.
Follow linked results to canonical pages before combining sources.

Misreading to avoid: An empty list means "no results for this lookup". An error message or unavailable source does not mean the data does not exist.

Source-specific next step

Go to the source page for other operations. Use the JSON template only after the HTML result shows that the parameters and source are correct.

AI agent? Read the complete navigation and programming guide.