Common Ground
Server Details
Pick one movie, show, book, artist or restaurant a whole group will enjoy, using Qloo taste data.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- theluckystrike/common-ground
- GitHub Stars
- 0
- Server Listing
- Common Ground MCP Server
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 2 tools
The two tools are clearly separated by output shape: find_common_ground returns a single pick from one category, while plan_night_out returns a three-stop itinerary (dinner, drinks, live music) in one city. Both share the same group-of-friends + Qloo taste-graph input model, so there is some conceptual overlap, but the wording makes the boundary clear enough to choose correctly.
Both names are snake_case verb-led phrases (find_common_ground, plan_night_out) that describe the action and its object. The only minor deviation is the differing number of words and the slightly metaphorical 'common ground', which is a bit less literal than the very explicit plan_night_out.
Two tools is thin for a recommendation server, and plan_night_out is essentially a composed, richer variant of find_common_ground. It is defensible because both are heavyweight, multi-parameter operations, but agents have almost no smaller building blocks to compose with.
The group-decision workflow (single pick and full evening plan) is covered, but the surface has dead ends: there is no way to list supported kinds/categories or cities, no follow-up to expand a single stop or refine a returned result, and no per-person or partial-group mode. Agents must get every parameter right in one call with no way to explore or verify.
Available Tools
2 toolsfind_common_groundFind common ground for a groupARead-onlyIdempotentInspect
Pick ONE movie, TV show, book, music artist, podcast, restaurant (kind=place, with location) or destination that a whole group will enjoy. Give each person's name and 2 to 5 things they love. Favourites can come from another category with tasteFrom, for example music artists to pick a restaurant, or any for a mix of artists, movies, TV shows and books. Uses Qloo's taste graph to score every candidate for every person and returns the option that makes the least happy person as happy as possible, with a reason per person.
Inputs: kind (required), members (required), location, tasteFrom.
plan_night_outPlan a night out for a group of friendsARead-onlyIdempotentInspect
Plan dinner, drinks and a live music venue in one city for 2 to 6 friends. Give the city and each friend's name with 1 to 4 music artists, movies, TV shows or books they love. Optionally give an exact night date as YYYY-MM-DD; Qloo then filters venues by that date's weekday only, not by exact opening times, travel or availability. Qloo's cross-domain taste graph turns that shared taste into real venues. Each stop is the venue whose least happy friend is happiest, and Qloo explainability names the favourites behind it, with the share of each friend's taste.
Inputs: city (required), members (required), date.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Plans a shared evening for two people: places both would enjoy, ranked by Qloo's taste graph.
Find movies and shows for tonight with current trends, mood, taste and streaming options.
Taste-based film, TV, book and game picks: similar in feel, taste reads, picks for two.
Help a group pick where to eat or travel: critic-backed restaurants, warm trips, one vote link.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to find movies, shows, books, music, podcasts, games, places, destinations, or brands that two people are likely to enjoy together. It resolves favorites to Qloo entities, compares taste profiles, ranks candidates for fairness, and returns explainable matches through five read-only tools.MIT
- AlicenseNot gradedqualityBmaintenanceEnables MCP clients to choose group-friendly options by generating per-person Qloo recommendations and rankings for a shared shortlist, then applying maximin fairness and local diet/allergy checks to surface a pick nobody ranks low.11 npmMIT
- AlicenseNot gradedqualityBmaintenancePlans an evening two people would both enjoy: give a city and a few favourites each, and Qloo's taste graph returns up to six places that rank well for both, each with the reason and in evening order.MIT
- AlicenseNot gradedqualityBmaintenanceEnables agents to discover and rank cultural places from supplied artists or films, clarify preferences through reversible pairwise choices, and rebuild recommendations using a shared API and decision engine.MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.