Search decisions

Full-text search across Swiss court judgments, court decisions and case law from entscheidsuche.ch. Covers decisions of the Federal Supreme Court (BGer, BGE) an

liveverifiedunknown~5s typical

mcpprobedentscheidsuche—schweizer

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-ch-entscheidsuche-mcp /search for me.

Input

Field	Type	Description
query	string	Full-text query. "..." for phrases, AND/OR/NOT for Boolean operators, * and ? for wildcard default "*"
language	string	Optional preferred language for highlighting/display (de, fr, it). If omitted, the first a
sort	string (relevance \| date \| scrapedate)	Optional sort order: 'relevance' \| 'date' (decision date) \| 'scrapedate'. Default is 'rele default "relevance"
size	integer	Results per page (1–100). default 20
search_after	string	Cursor from `next_cursor` of the previous response.
decision_date_from	string	Lower bound for the decision date (YYYY-MM-DD, inclusive).
decision_date_to	string	Upper bound for the decision date (YYYY-MM-DD, inclusive).
scrape_date_from	string	Lower bound for the scrape date (YYYY-MM-DD, inclusive).
scrape_date_to	string	Upper bound for the scrape date (YYYY-MM-DD, inclusive).
hierarchy	string	List of hierarchy IDs (canton/court/chamber). Multiple IDs are combined with OR; with othe
language_filter	string	Filter by document language(s).
include_aggregations	boolean	Include aggregations (distributions over the results). default false

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-ch-entscheidsuche-mcp","endpoint":"/search","input":{"query":"*","sort":"relevance","size":20,"include_aggregations":false}}'

cli

npx cracked-ai run -p mcp-ch-entscheidsuche-mcp -e /search -i '{"query":"*","sort":"relevance","size":20,"include_aggregations":false}'

mcp

run_tool({ provider: "mcp-ch-entscheidsuche-mcp", endpoint: "/search", input: {"query":"*","sort":"relevance","size":20,"include_aggregations":false} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.