OpenApps MCP server

• Registry namecom.theopenapps/openappsverified in the MCP registry

The registry name sits under theopenapps.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

• Stateanswerschecked 2026-10-28
• Tools14
• Server calls itselfopenapps
• Server version91d12c87-ba31-44f7-af35-467e24ac2166
• Protocol version2025-06-18

On 2026-10-28 the server answered initialize and tools/list without a sign-in. First answer in 65 ms.

• Its tools do every use case we checked: Check daily rank moves for apps in each store; Spot new app launches the day they appear.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

• INITE MCP score66 of 100answers 100 · does the job 100 · definitions 86 · context 100 · connects 100 · lowered by the warning signs below
• What an agent can do with it2 of 2 use cases
• Answered100% of days7 of 7 daily checks
• Median first answer886 ms
• Space in an agent's context≈ 2,739 tokensthe definitions of all its tools (14), which an agent carries on every turn
• Tool definitionsB · 86 of 100
• Typed results2 of 14 toolsdeclared outputSchema: results come typed, read once

• Parameters without a description: in 12 of 14

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

• changes_list

  Use this when the user asks what changed in a store: chart moves, featuring, launches, delistings, metric moves or MCP changes, newest first with counts per type. Do not use for a chart's full current order; use charts_get.

• new_list

  Use this when the user asks which apps are new in a store since a time, by launch date or else the day each went live. Do not use for rank or toolset changes; use changes_list.

• stores_get

  Use this when you need stores, app counts, countries, freshness, charts or metrics. Pass store for chart names. guide returns a playbook's steps: weekly_brief (a store's week), competitor_watch (compare apps), audit_my_server (an MCP…

• charts_get

  Use this when you need a chart now or at a past time, with movement and 1-based positions. Do not use for a listing across charts; use listings_profile.

• leaderboards_get

  Use this when the user asks which apps lead a store by a published number. claude: installs, popularity_score, trending_score, gh_stars; cursor: dir_installs, dir_stars. Do not use for chart positions; use charts_get.

• scores_list

  Use this when the user asks which apps are popular in a store over the last month: OpenApps Rank, a 28-day aggregate of its popularity charts, with tier and momentum. Do not use for one day's chart order (charts_get) or published counts…

• listings_search

  Use this when you need an app's ref from its name, its publisher, a brand it works with or what it does ("create jira ticket"); one row per app, best match first. Do not use for popularity or rankings; use charts_get.

• featuring_history

  Use this when you need when and where apps were featured, per app or home shelf, over at most 30 days. Do not use for current shelf order; use charts_get.

• listings_profile

  Use this when you need a listing profile, or up to 5 (Pro): its app, country availability, ranks, metrics and MCP tools. Stale or refused countries read unknown. Do not use for discovery; use listings_search.

• mcp_toolset_get

  Use this when you need a tracked app or server's stored MCP tools or changes. at selects a past snapshot; diff_from compares versions. Do not use for arbitrary URLs; only tracked servers.

• market_get

  Use this when you need store sizes, weekly launches, app kinds or MCP ecosystem statistics. Do not use for app profiles or rankings.

• search

  Use this when ChatGPT deep research needs listings, apps, tags (query tag:<id>), charts or tracked MCP URLs. Returns typed ids for fetch. Do not use for rankings or events; use charts_get or changes_list.

• fetch

  Use this when ChatGPT deep research needs a document for an id returned by search: listing, app, tag, chart or MCP toolset. Do not use for ids outside search results.

• feedback_send

  Use this when OpenApps' tools or data fail the user: an error, data the user says is wrong (such as an app bound to the wrong listing), a missing capability, or something confusing. Ask the user before sending a report they did not ask…