Booli via fpx (no MCP)

Booli fronts www.booli.se — including /graphql — with a Cloudflare managed challenge that 403s any plain curl/Node request. fpx routes the request through the user's own signed-in browser tab (the ContextMint Bridge extension), which has already cleared the challenge, so the same anonymous GraphQL query succeeds. No Booli login is needed — just a normal open tab.

This is the same data the booli_* MCP tools return (money in SEK, areas in m², rooms as a number), reached with one CLI call instead of a running server.

One-time setup

npm install -g @fetchproxy/cli           # provides `fpx`
fpx profile add booli --domain booli.se  # only the fetch capability is needed
fpx pair -p booli                        # prints a pair code → approve in ContextMint Bridge

Requirements: the ContextMint Bridge browser extension installed (from https://github.com/nullnet-app/contextmint-bridge/releases — Chrome: load the chrome zip unpacked; Safari isn't available yet — use Chrome for now), with an open www.booli.se tab, and its Chrome Site access allowing booli.se. ContextMint Bridge is the fetchproxy extension renamed, same maintainer; source at https://github.com/nullnet-app/contextmint-bridge (build it, or verify a release zip with its published .sha256). Pairing persists — after the first approval every later fpx call reuses it.

Core call

Every query is a POST of {"query": "...", "variables": {...}} to the GraphQL endpoint. Write the body to a file (a heredoc avoids shell-quoting the GraphQL string), then send it so stdout is the GraphQL JSON, ready for jq:

cat > /tmp/bq.json <<'JSON'
{ "query": "…", "variables": { … } }
JSON
fpx post-json 'https://www.booli.se/graphql' @/tmp/bq.json -p booli | jq '.data'

Ready-to-run query bodies (area lookup, for-sale search, sold search, property detail) with jq recipes are in references/graphql-queries.md. Exhaustive field lists live in the repo at src/graphql.ts — the operations here are compact, live-verified subsets.

The one rule: resolve the area first

Booli searches take a numeric area id (areaId), never a place name. Always resolve the name first, take the id, then search:

printf '{"query":"query($s:String!){areaSuggestionSearch(search:$s){suggestions{id displayName}}}","variables":{"s":"Nacka"}}' > /tmp/a.json
fpx post-json 'https://www.booli.se/graphql' @/tmp/a.json -p booli \
  | jq -r '.data.areaSuggestionSearch.suggestions[] | "\(.id)\t\(.displayName)"'
# 76      Nacka kommun

Pass that id as the search input's areaId (a string). The SearchRequest input needs areaId, page (1-based), ascending, excludeAncestors: true, facets: [], filters: [{key,value}], and sort ("" = newest). Filters reuse the same keys the MCP does (minRooms, maxListPrice, objectType, …).

Exit codes (fetch verbs)

• 0 — success (a GraphQL errors array can still ride in a 0 body; check jq '.errors // empty').
• 2 — bridge unavailable: extension not connected or pairing pending → run fpx pair -p booli, confirm a booli tab is open.
• 3 — bot wall: the tab hasn't cleared Cloudflare → open/refresh a www.booli.se tab and retry.
• 4 — upstream non-2xx from Booli.

Notes

• Anonymous reads only — the same public GraphQL surface as the MCP; no account data. Stay within booli.se's terms.
• fpx health -p booli shows bridge connection state when a call fails.
• Property detail is keyed by residence id — the number in a /bostad/<id> URL, which is a search result's url field, not booliId.
• This project is developed and maintained by AI (Claude).