
StockDrifts MCP server
Which MCP servers are StockDrifts's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-10-26
- Registry nameio.stockdrifts/mcpverified in the MCP registry
The registry name sits under stockdrifts.io. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-10-26
- Tools32
- Server calls itselfstockdrifts
- Server version1.29.0
- Protocol version2025-06-18
On 2026-10-26 the server answered initialize and tools/list without a sign-in. First answer in 517 ms.
- Its tools do Follow insider buying in Korean and Japanese markets, but not Review SEC filings and earnings call transcripts for a stock; Compare analyst ratings and consensus price targets.
- Checked on 2 of the 7 days it takes to judge reliability.
Once connected
- Parameters are untyped or undescribed in 32 of 32 tools: check arguments before calling.
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 scoreafter a weeka score needs 7 daily checks: 2 so far
- What an agent can do with it1 of 3 use cases
- Answeredmeasuring2 of the 7 daily checks it takes to print a share
- Median first answer589 ms
- Space in an agent's context≈ 6,457 tokensthe definitions of all its tools (32), which an agent carries on every turn
- Tool definitionsC · 67 of 100
- Typed results0 of 32 toolsdeclared outputSchema: results come typed, read once
- Parameters without a description: in 32 of 32
- Not saying whether it only reads or can destroy: 32 of 32
What an agent can do with it
- cannot · Review SEC filings and earnings call transcripts for a stock
- cannot · Compare analyst ratings and consensus price targets
- can do · Follow insider buying in Korean and Japanese markets
Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?
Worth a look
None of the five warning signs in these definitions.
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 →
list_insider_trades_krInsider trades for one KR ticker
get_insider_summary_krPre-aggregated insider summary for one KR ticker
list_recent_trades_krRecent KR insider trades across all tickers
list_top_buys_krTop KR tickers by aggregated insider buying over a window
list_top_sells_krTop KR tickers by aggregated insider selling over a window
list_opportunities_krRecent KR open-market BUYs whose price hasn't moved much yet
list_plans_krKR pre-disclosure (D005) plans — announced but not yet executed trades
get_plan_executions_krCross-reference one D005 plan against its actual executions
search_companies_krSearch KR companies by stock_code, Korean name, or English name
list_leaderboard_krTop KR insiders by historical avg return on their buys
list_insider_filings_jpShareholding-report filings for one JP ticker
get_insider_summary_jpPre-aggregated shareholding summary for one JP ticker
list_recent_filings_jpRecent JP filings across all tickers
list_top_buys_jpTop JP tickers by aggregated insider buying over a window
list_top_sells_jpTop JP tickers by aggregated insider selling over a window
list_investor_filings_jpAll JP filings made by one investor (EDINET code)
list_opportunities_jpRecent JP BUY/INITIAL filings whose price hasn't moved much yet
list_leaderboard_jpTop JP investors by star rating + avg return
list_convergence_jpJP tickers with multiple distinct investors filing on the same name
search_companies_jpSearch JP companies by stock_code, Japanese name, or English name
search_investors_usFuzzy-search investors by name; resolves to CIKs ordered by AUM Case-insensitive partial-name match against the latest-holdings view. Returns the largest funds first so the most likely match is at the top. Use this to resolve a manager…
screen_investor_sector_exposure_usScreen investors by sector concentration (e.g. top biotech specialists) Ranks 13F filers by the share of their portfolio in one sector, computed from sector-tagged top-50 holdings. `tier` buckets: specialist (>=60%), heavy (30-60%)…
screen_investor_catalyst_record_usScreen funds by their track record on binary events (any sector) Ranks 13F filers by how their positions resolved through binary events — single-day moves >= ~25%, price-defined, NOT verified cause labels. In biotech these are trial…
get_investor_holdings_usLatest 13F holdings for one investor (or for a specific period)
list_investor_periods_usAll 13F periods on file for one investor
get_investor_changes_usQoQ position diff for one investor between two periods
get_investor_sector_exposure_usSector-exposure history for one investor (newest quarter first) Per-quarter portfolio composition by sector for one 13F filer, including the biotech tier and top biotech positions. Use it to see whether a fund was already a sector…
get_investor_catalyst_record_usOne fund's binary-event track record per sector, with best/worst calls The fund's record of holding names through binary events (price-defined single-day moves — readouts/FDA/buyouts in biotech, earnings/M&A elsewhere; not verified cause…
list_ticker_holders_usInvestors with `ticker` in their top-50 positions, sorted by position size Returns 13F filers that hold `ticker` **as one of their 50 largest positions**, scoped to the most recent reporting period by default. Only top-50 positions per…
get_kpiLatest normalized KPI snapshot for one ticker
screen_insidersScreen insider activity by country, date, value, role, and ticker Screens one market per call, selected by `country`: 'KR' returns individual insider transactions, 'JP' returns 5%+ shareholding reports. The response echoes `country`…
screen_holdings_usScreen 13F holdings by period, ticker, value, and rating; returns each matched fund with only the position(s) that matched Results are wrapped in an envelope: `results` holds up to `limit` funds (max 500), `total_count` is the full…
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"stockdrifts": {
"type": "http",
"url": "https://mcp.stockdrifts.io/mcp/"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http stockdrifts https://mcp.stockdrifts.io/mcp/claude.ai, ChatGPT and other clients: add https://mcp.stockdrifts.io/mcp/ as a custom connector in the settings.
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Cookies
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.