Skip to main content
Glama
Tickerbot MCP Server

Tickerbot MCP Server

Official
by tickerbot

tickerbot_get_history

Retrieve time-series signals or fundamental data for one ticker or up to 50 tickers on a shared grid, paged backward, to chart history or read a point-in-time state.

Instructions

Returns the signals of one ticker or a comma list of up to 50 in the path (a screen's matches, say) over time, all-time: up to 25 signals in select, or * for every signal the grain stores, every ticker on one shared grid. series is keyed by symbol, one flat row per grain step. Paged backward: pass next_cursor back as cursor with the same params. For when a yes/no signal flipped, use tickerbot_query_events with kind=signal.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "ticker": "<ticker>"
    },
    "name": "tickerbot_get_history"
  }
}

Input Schema

NameRequiredDescriptionDefault
toNoWindow end, `YYYY-MM-DD` or ISO. A bare date covers that whole day. A timestamp is included until 2027-01-12 and exclusive from then, as everywhere else (Deprecations). At `1m`/`1h` a bare date on `from` or `to` is each ticker's market day: the ET calendar day for stocks and ETFs, the UTC day for crypto, FX and rates, ticker by ticker in a mixed list. See Time windows.
asofNoPoint-in-time read: one result per ticker (the state at that instant) rather than a range. `YYYY-MM-DD` or a full ISO timestamp, the same meaning `asof` carries on `/v2/tickers/{ticker}`, `/v2/scan` and `/v2/signals/{signal}`. Combined with `from`/`to` or `cursor` it is a 400 (a point and a window contradict), and `limit` has no meaning under it. It also resolves which company held the symbol at that instant: a ticker that changed hands returns the state of whoever traded it then, so `/v2/tickers/SHLD/history?asof=2010-06-30` returns Sears Holdings' price and `asof=2026-01-01` returns the Global X defense ETF. Returns the most recent result at or before the instant, so a date inside a trading gap gives the last result before it; on the `1d` and `1w` grids a timestamp returns the previous session's result, never that day's close. Also: `next_cursor` is always `null` (every requested ticker is on the one page). A ticker with no data on the grid at or before the instant (outside the minute tier, say) comes back as an empty array with `_meta.coverage[ticker].rows: 0`; a 404 means only that we do not track the symbol. A bare date on the `1m`/`1h` grid is accepted and means the last result of that day: the ET calendar day for stocks and ETFs, the UTC day for crypto, FX and rates (the state reads 400 that combination; this grid read does not). The look-back is the grid's window: `1d` walks back without limit, `1h` 60 days and `1m` 7 days before the instant. At `grain=1q` a quarter is visible from its `available_at`, when the market could first trade on it: 9:30 AM ET on the report date for a pre-market earnings release, else 9:30 AM ET of the next session; statements the session after filing. A bare date means that day's 4 PM ET close, so you get the latest quarter that was public knowledge at the instant, with restatements after it excluded: `/v2/tickers/AAPL/history?grain=1q&asof=2026-07-30` returns 2026-Q1, `asof=2026-07-31T13:30:00Z` 2026-Q2 (the June quarter was reported after the close on 2026-07-30). See availability. At `grain=1y` a fiscal year is visible from its `available_at`: 9:30 AM ET of the first session after the original annual report was filed. A bare date means that day's 4 PM ET close: `/v2/tickers/AAPL/history?grain=1y&asof=2025-10-31` returns fiscal 2024, `asof=2025-11-03` fiscal 2025 (the 10-K was filed after the close on Friday 2025-10-31). Response time: under 1.5 s cold, under 0.6 s warm.
fromNoWindow start, `YYYY-MM-DD` or ISO: a bare date starts at the beginning of that day, a timestamp is included. At `1q` and `1y` it bounds the fiscal period end. Intraday requests default to a recent window (`1m`: 7 days, `1h`: 60 days); the cursor keeps walking further back window-by-window, or pass `from` to widen it up front.
grainNoGrid granularity. `1w` resamples the daily grain weekly (Monday-keyed); `1q` is the fiscal-quarter grid; `1y` is the fiscal-year grid of annual statements. This is the grid: there is no `auto` here and nothing blends; under `asof` you get the last result on this grid, and a signal the grid does not carry is a `400`. Values: 1m: Minute. One result per minute, for the signals stored at one minute. 1h: Hour. One result per hour, for the signals stored hourly. 1d: Day. One result per day, for the signals stored daily: the most signals and the longest history. 1w: Week. The daily results resampled by week, keyed by each week’s Monday. A signal takes the week’s last value; price bars aggregate (open first, high highest, low lowest, close last, volume summed). 1q: Fiscal quarter. Quarterly fundamentals on the company’s fiscal calendar, keyed by quarter such as `2026-Q2`. Name them in `select`; they come in a request of their own, apart from daily and intraday signals. 1y: Fiscal year. Annual statement figures on the company’s fiscal calendar, keyed by fiscal year such as `2025`.1d
limitNoGrid steps per page. Max 1000; an over-cap `limit` is clamped to 1000. Separately, a page holds at most 25,000 points in all: tickers × `limit` over 25,000 is a 400 (50 tickers take `limit` up to 500; 25 take the full 1,000). With `select=*` the page holds at most 1,000 points: over it, an explicit `limit` is a 400. Without `limit`, a `*` page holds 252 points (one ticker gets 252, 5 tickers get 50).
cursorNoContinue paging back: the previous response's `next_cursor`, unchanged. Send it with the same params as the first page (`limit` may change); a page whose params differ is a 400 naming the param. On `1m` to `1w` it carries each ticker's own position, so every ticker continues from its own oldest row: nothing is skipped or repeated, even when the tickers trade on different calendars (a stock and a crypto pair). On `1q`/`1y` it carries the shared fiscal label, so every ticker continues from the same period.
selectNoUp to 25 signals: built-in signals, your custom signals and OHLCV names, freely mixed. `*` is every signal the grain stores (at `1q` and `1y`, every quarterly or annual field), with no 25-signal cap; its page holds at most 1,000 points (tickers × `limit`), see `limit`. `*` returns the same rows as a named list of stored signals, and its `vwap` is the stored value (a named `vwap` is read from the bars). A signal sent beside `*` that `*` already holds adds nothing; an OHLCV name or one of your custom signals beside it is added; an unknown name is a 400. Omitted → the ticker-history default set (price, change_1d_pct, relative_volume, market_cap), intersected with what the grain carries. At `1q`, `select` is required and quarterly-only; at `1y`, required and annual-only (the list is under Notes).
tickerYesOne symbol, or a comma-separated list of up to 50 for a response keyed by symbol, on one aligned grid. A page holds at most 25,000 points in all (tickers × `limit`), so 50 tickers page at up to 500 points each. Case-insensitive. Equities are bare symbols (`NVDA`); every other class carries a prefix: rates (`R:SOFR`), crypto (`X:BTCUSD`), fx (`X:EURUSD`). Bare `BTC`/`ETH` are US-listed ETFs, not spot crypto. See Tickers.

Output Schema

NameRequiredDescriptionDefault
_metaYesThis endpoint's own keys: per-ticker `coverage` (`rows`, and `missing_bar_cells` when a bar signal is read on a state spine), `non_trading_days_dropped` and on an intraday range read `from_defaulted` with a `note`. Shared keys: see the `_meta` reference.
as_ofYesServer time this response was assembled (ISO 8601); on an as-of read, the instant you asked for (As-of).
countYesThe rows in `series` on this page, across all tickers.
queryYesThe request as it was read, with each param's effective value: `ticker` (a list, even for one symbol), `from`, `to`, `asof`, `grain`, `select` (the signals served; `["*"]` when you sent `*`) and `limit`.
seriesYesThe rows, keyed by ticker (one ticker or a comma list alike). Each row is a flat object, oldest first: `t` (the grid step: a date at `1d`/`1w`, an ISO instant at `1m`/`1h`, and at `1q`/`1y` a fiscal label, `2026-Q2` or `2025`: one of two times in the data that are not ISO 8601, with a bar's `t`) plus the signals you selected.
not_foundNoWith a comma list only: the requested symbols we do not track, in request order; an empty array when every symbol was found; every symbol when none is tracked (still a 200). Until 2027-01-12 each also keeps an empty entry in `series` and `_meta.history_revised_at`; from then it is listed here only.
_advisoriesNoPresent only when the response carried a warning: `deprecated` (since, sunset_on, successor, warning; on a deprecated tool, successor names the tool to call), `rate_limit` (warning, retry_after_seconds, limit, remaining, reset_epoch_s) or `monthly_quota` (warning, limit, remaining, reset_epoch_s).
next_cursorYesOpaque token for the next page; `null` once every ticker is finished. Pass it back as `cursor`.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.4.1

TDQS

A4.3/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does disclose real behavior: results are keyed by symbol and aligned on one shared grid, pages walk backward only, and a signal absent from the chosen grain is rejected rather than blended. It omits auth requirements, rate limits, and error/response-shape behavior beyond paging, which keeps it short of a 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Four sentences, front-loaded with the purpose, and each subsequent sentence adds paging or routing information. The first sentence is overloaded, chaining ticker count, select cap, and grid alignment, and some content duplicates the rich schema, but there is little outright waste.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

This is a complex 8-parameter read with an output schema, and the description covers the parts an agent cannot infer: grid alignment across tickers, one-way cursor paging, and the events alternative. Return values and per-parameter details are correctly delegated to the output schema and the 100%-covered input schema.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so baseline is 3, but the description adds meaning the schema does not: series is 'keyed by symbol, one flat row per grain step' and every ticker sits on one shared grid, which tells the agent how the output is shaped. The remaining per-parameter caveats (limit caps, asof 400s) largely restate the already descriptive schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The opening sentence gives a specific verb (Returns), a specific resource (signals), and the scope (one ticker or a comma list up to 50, over time, all-time), plus the select cap and the '*' variant. It also names a sibling it is not, routing event-flip questions to tickerbot_query_events with kind=signal, so an agent can distinguish it without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It states concrete operational context: only backward paging, next_cursor replayed as cursor with the same params, and the explicit alternative (tickerbot_query_events with kind=signal) for flip timing. It never mentions the other adjacent siblings such as tickerbot_get_signal or tickerbot_get_bars, so the alternatives coverage is partial rather than complete.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.