Skip to content
Migrate your GPTCheck my InvokeRank
InvokeIndex

The tools agents can call

150,674 unique MCP servers across ten major directories and registries, each with its sources and history.

Each server counts once, however many directories list it. Glama, PulseMCP and mcp.so listings arrive through MCP Toplist’s open dataset, used under CC BY 4.0. 14,013 listings from bulk publishers wait for review.

Check my InvokeRankUpdated How we collect
  • Directory listings58,190Collected from seven directories, before grouping
  • Listed in two or more32,91722% of unique servers
  • Remote MCP servers26,499With a public endpoint an agent can call
  • Changes, last 30 days50,315Typed diffs between observations

PulseMCP (22,396)

Plus 202 listings grouped under these serversPage 182 of 934
Inside the index

Where the servers are listed

Each directory is a shelf. See how many shelves each server sits on, how large each shelf is, and what moved this month.

Updated

How many shelves

Servers by the number of directories listing them

78%of servers are listed in one directory only. 32,917 are listed in two or more.

Directories listing the server. Copies grouped under a server are not counted again.

The shelves

Unique servers listed in each directory
  1. Glama97,241
  2. MCP Registry38,642
  3. PulseMCP22,396
  4. mcp.so18,556
  5. Smithery12,483
  6. ChatGPT5,983
  7. Claude5,052
  8. Cursor424
  9. GitHub388
  10. Claude Code337

Collected by us Via MCP Toplist’s dataset

A server listed in several directories counts once in each. Pick a shelf to browse it.

What changed

Typed diffs between observations, last 30 days
New listings 6,535Left a directory 361Tool changes 34,933Listing details 8,486The busiest day was 9 Oct, with 9,191 changes.
Changes by day, last 30 days (days with changes)
DayNew listingsLeft a directoryTool changesListing details
2 Oct26540140
3 Oct685203212969
4 Oct51563272445
5 Oct942293959626
6 Oct6461559401706
7 Oct7372250981736
8 Oct7062415021865
9 Oct16671759341573
10 Oct37272497426

Latest changes

RSS feed
  1. DataPorium Official MCP Registry screen_ma_deals: input schema changed (+by, +company, +filters, +query, +view)
  2. DataPorium Official MCP Registry screen_ma_deals: description changed (+"Private fundraising (official public data, 2009 on, updated daily): private (unregistered) offerings of US and foreign companies and funds, one notice per filing, each offering shown once at its newest filing by default. view "private_fundraising": rounds newest first (sort_by "value" = largest amount sold): issuer, industry group, state or country, offering amount, amount sold and remaining (US dollars, as stated), investors, minimum investment, first sale date, exemptions (506b, 506c, 504, 3c1, 3c7, ...), fund or operating company, ticker if listed. query = part of the issuer name or its company id; date_from/date_to filter the filing date (YYYY-MM-DD, YYYY-MM, YYYYQn or YYYY); filters e.g. {"industry": "Biotechnology", "state": "CA", "kind": "operating", "min_sold": 10000000, "exemption": "506c", "fund_type": "venture_capital_fund", "revenue_range": "1m_to_5m", "ticker": "NVDA", "listed": true, "filings": "all", "sort": "investors"}; list_filters('screen_ma_deals') shows all. view "private_fundraising_company": one issuer (company = slug or company id from the rows) with its offerings and every filing over time. view "private_fundraising_totals": totals by quarter (default; by), year, month, industry, category, sector, state, country, fund_type, exemption, revenue_range, entity_type or none: new offerings, issuers, funds vs operating companies, amount sold, offering amount and medians; filters as above plus {"basis": "latest"|"all"}. About half of all filings are pooled funds and a few large funds dominate the sums: use {"kind": "operating"} for companies only. These are notices of offerings, not verified results.")
  3. DataPorium Official MCP Registry get_short_interest: input schema changed
  4. DataPorium Official MCP Registry get_short_interest: description changed (+"view "fails_to_deliver" (added 10/09/2026): shares not delivered to the buyer by the settlement date, per settlement day since 2017, published twice a month (1st to 14th, 15th to month end). With one symbol: daily rows (quantity = open balance of shares not delivered that day, price of the day before, value = quantity x price, newest first; date_from/date_to), totals per half-month (days with fails, average, largest day, longest run), the largest day and the newest half-month. Without a symbol: the largest fails of the newest half-month (settlement_date = a date inside another half-month); sort max_value (default), avg_value, total_value, max_quantity, avg_quantity, days_with_fails or longest_run; filters e.g. {"listed_only": true, "flagged_only": true, "min_value": 1000000}. flagged = at least 10,000 shares on 5 or more settlement days in a row (a simple screen, not an official list). Daily numbers are open balances, not new fails; a fail can have ordinary causes and does not by itself show improper short selling: say so when you use it.")
  5. DataPorium Official MCP Registry get_salaries: input schema changed (+college, +field)
  6. DataPorium Official MCP Registry get_salaries: description changed (+"College costs and graduate pay (official public data, US colleges operating now; added 10/09/2026): view "colleges": colleges ranked or searched: query (name words) or state; filters e.g. {"ownership": "public", "level": "four_year", "max_net_price": 20000, "max_admission_rate": 0.3, "size": "large", "type": "hbcu", "city": "Austin", "county": "Travis"}; sort enrollment (default), net_price, tuition_in_state, admission_rate, earnings_10yr, median_debt, debt_to_earnings, completion_rate or repayment_rate (rankings leave out colleges with fewer than 500 undergraduates unless filters {"min_enrollment": 0}). Net price = yearly cost after grants; earnings = median of former students 10 years after they started. view "college": one college (college = college_id or slug from view colleges, or query: the best name match): every measure, its biggest and best paid programs, tuition and net price by year, its earnings rank, medians to compare. view "college_programs": programs (field of study x credential) with graduates, median debt and median earnings 1 and 4 years after completion: one college's programs (college), or a field across colleges ranked by earnings (field = 4-digit code such as 1107 computer science or 5138 nursing, or a 2-digit family such as 11; or query = words of the field title); filters e.g. {"credential": "bachelor", "ownership": "public"}; sort earnings_1yr, earnings_4yr, median_debt, debt_to_earnings or completions. view "college_fields": fields of study (majors) ranked by the median earnings of their programs 1 year after completion ("which majors pay the most?"): filters e.g. {"credential": "bachelor"}; field or query narrow it. Small programs' earnings and debt are withheld for privacy (null). Say these are medians of graduates who work.")
  7. DataPorium Official MCP Registry get_recalls: input schema changed (+carrier, +facility)
  8. DataPorium Official MCP Registry get_recalls: description changed (+"ratings," +"complaints, trucking carrier safety, and workplace safety, environmental compliance and greenhouse gas emissions of facilities and listed companies" +"view "carriers": active US trucking carriers with 10+ power units, largest fleet first: power units, drivers, crashes over 24 months (fatal, injury, tow-away), crash rate per 100 power units, roadside inspections and driver and vehicle out-of-service rates, safety rating, ticker. Find by query (name or carrier number), ticker (e.g. JBHT, KNX, WERN, ODFL, SAIA), filters e.g. {"state": "TX", "kind": "passenger", "operation": "interstate", "safety_rating": "conditional", "min_power_units": 100}; sort power_units, drivers, crashes, fatal_crashes, crash_rate, vehicle_oos_rate, driver_oos_rate (carriers with 20+ inspections) or inspections. view "carrier": one carrier in full (carrier = carrier number or slug from view carriers) with its fleet rank and the averages of all trucking carriers, of its state and of all roadside inspections. view "trucking_stats": all carriers (also small ones) by state (default, last 12 months), month or year (by): active carriers, power units, drivers and fleet sizes (monthly snapshot), new carriers, crashes and deaths (where they happened), inspections and out-of-service rates; date_from/date_to as YYYY-MM, filters e.g. {"state": "TX"}; sort (by state) e.g. crashes or vehicle_oos_rate. Crashes do not say who caused them. Workplace safety, environmental compliance and emissions (2010 on; counts and money, no personal data): view "facilities": facilities with their totals for the last 5 years (date_from/date_to as years): workplace safety inspections, violations (serious = serious, willful or repeat) and penalties; environmental (air, water, waste) inspections, violations, formal enforcement actions, penalties and quarters in noncompliance; newest yearly greenhouse gas emissions. Find by query (part of a name), ticker (e.g. XOM, DUK, TSN, WMT, AMZN), filters e.g. {"state": "TX", "county": "Harris", "zip": "77506", "lat": 29.7, "lon": -95.0, "radius_miles": 5, "kind": "environmental", "program": "air", "naics": "324"}; sort penalties (default), violations, serious_violations, inspections, enforcement_actions, emissions or distance. view "facility": one facility in full (facility = facility_id from view facilities): every year of both kinds and of emissions, owners and listed company. view "compliance_companies": with ticker, a listed company's facilities by year and kind (inspections, violations,")

Coverage

What the index resolves and probes
Brands resolved
25,077
Brands on two or more hosts
5,631
Held for review
14,013 listings
Tools listed by probe
10,518 servers

Where are you listed?

Paste your MCP server, plugin or company URL to see your listings, tools and scan findings in one place.

Check my InvokeRank Scan a server