# io.github.cyanheads/exchange-rates-mcp-server

> Convert currencies, get FX rates, and query historical ECB exchange rate data.

- name: io.github.cyanheads/exchange-rates-mcp-server
- version: 0.5.0
- connection class: R0 (autonomous)
- trust: 94/100
- quality: 86/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/io.github.cyanheads%2Fexchange-rates-mcp-server
- tools: https://api.protogrid.dev/v1/servers/io.github.cyanheads%2Fexchange-rates-mcp-server/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "exchange-rates-mcp-server": {
      "type": "http",
      "url": "https://exchange-rates.caseyjhand.com/mcp"
    }
  }
}
```


## Trust

- hygiene: 100
- liveness: 100
- freshness: 100
- provenance: 80
- drivers: +repository +namespace-matches-repo +reachable +uptime-100% +updated-1d-ago
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 86/100 (good)

### protocol: 88 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- warn `protocol.list_ttl`: tools/list declares ttlMs 0, so clients re-fetch it on every use

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 79 (weight 30)

- pass `tools.descriptions`: every tool has a description
- warn `tools.description_length`: 1 tools have descriptions over 1,024 characters, which crowds the context
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 7 of 7 tools declare readOnlyHint or destructiveHint
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: fx_get_rates (destructiveHint); fx_get_timeseries (destructiveHint); fx_dataframe_describe (destructiveHint); …
- pass `tools.token_cost`: about 2,185 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 100 (weight 15)

- pass `stability.changes`: 3 tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: deps.dev does not know @cyanheads/exchange-rates-mcp-server@0.5.0, @cyanheads/exchange-rates-mcp-server@0.5.0

- tools: 7, about 2,185 tokens to load them all
- tool set hash: f557f0e08753a8fdb11199010117dc02
- tools last changed: 2026-10-20T11:15:06.940Z
- badge: [![protogrid quality](https://protogrid.dev/badge/io.github.cyanheads/exchange-rates-mcp-server.svg)](https://protogrid.dev/servers/io.github.cyanheads/exchange-rates-mcp-server)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-20 `fx_dataframe_query` changed (description)
  - before: Run a read-only SQL SELECT against DataCanvas tables staged by fx_get_timeseries. Supports aggregations, GROUP BY, window functions, and JOINs across multiple registered tables. Run fx_dataframe_describe first to discover table names and column schemas. Requires DataCanvas (CANVAS_PROVIDER_TYPE=duckdb) — without it this tool is not listed at all and fx_get_timeseries returns every range inline.
  - after: Run a read-only SQL SELECT against DataCanvas tables staged by fx_get_timeseries. Supports aggregations, GROUP BY, window functions, and JOINs across multiple registered tables. Run fx_dataframe_describe first to discover table names and column schemas. Requires DataCanvas, which is on by default — a deployment that turns it off does not list this tool at all, and fx_get_timeseries returns every range inline there.
- 2026-10-20 `fx_dataframe_describe` changed (description)
  - before: List tables and columns staged on a DataCanvas from a prior fx_get_timeseries call. Required first step before fx_dataframe_query — use it to discover table names and column schemas. Requires DataCanvas (CANVAS_PROVIDER_TYPE=duckdb) — without it this tool is not listed at all and fx_get_timeseries returns every range inline.
  - after: List tables and columns staged on a DataCanvas from a prior fx_get_timeseries call. Required first step before fx_dataframe_query — use it to discover table names and column schemas. Requires DataCanvas, which is on by default — a deployment that turns it off does not list this tool at all, and fx_get_timeseries returns every range inline there.
- 2026-10-20 `fx_get_timeseries` changed (description, outputSchema)
  - before: Get historical daily exchange rates for a currency pair over a date range. ECB publishes on business days only — weekends and holidays produce no entry, and no date outside the requested range is ever returned, so a range covering only non-publication days comes back with an empty rates map and a notice explaining why. A same-currency pair returns a rate of 1 on each publication day in the range. Inline results are returned as a date→rate map paged at 500 publication days: rate_count is always the total for the requested range, and when a page is cut short the response carries truncated=true and next_start_date — call again with start_date set to next_start_date and the same end_date for the next page. When DataCanvas is enabled (CANVAS_PROVIDER_TYPE=duckdb) long ranges (>90 days by default) spill to it instead: the response carries spilled=true, a canvas_id, and a table_name — call fx_dataframe_describe to inspect the staged table, then fx_dataframe_query to run SQL against it. Without DataCanvas long ranges are paged inline (spilled=false) and the notice says so.
  - after: Get historical daily exchange rates for a currency pair over a date range. ECB publishes on business days only — weekends and holidays produce no entry, and no date outside the requested range is ever returned, so a range covering only non-publication days comes back with an empty rates map and a notice explaining why. A same-currency pair returns a rate of 1 on each publication day in the range. Inline results are returned as a date→rate map paged at 500 publication days: rate_count is always the total for the requested range, and when a page is cut short the response carries truncated=true and next_start_date — call again with start_date set to next_start_date and the same end_date for the next page. Long ranges (>90 days by default) are considered for DataCanvas staging: when the series exceeds the 40,000-character preview budget, it is staged and the response carries spilled=true, a canvas_id, and a table_name — call fx_dataframe_describe to inspect the staged table, then fx_dataframe_query to run SQL against it. On a deployment with DataCanvas turned off, long ranges are paged inline (spilled=false) and the notice says so.

Full history: https://api.protogrid.dev/v1/servers/io.github.cyanheads%2Fexchange-rates-mcp-server/changes

## Dependencies

- npm @cyanheads/exchange-rates-mcp-server@0.5.0: unknown
- npm @cyanheads/exchange-rates-mcp-server@0.5.0: unknown

Dependency graphs from deps.dev; advisories from OSV.dev, including the GitHub Advisory Database and the PyPI Advisory Database; all CC BY 4.0.

## Remotes

- https://exchange-rates.caseyjhand.com/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

- npm @cyanheads/exchange-rates-mcp-server@0.5.0
- npm @cyanheads/exchange-rates-mcp-server@0.5.0

## Tools (7)

- `fx_convert_currency`: Convert an amount between any two currencies at the latest or a historical rate. Returns the converted amount, the rate used, the actual rate date, and whether the date was snapped from a weekend/holiday to the prior business day. Cross-rates are triangulated through EUR automatically.
- `fx_dataframe_describe`: List tables and columns staged on a DataCanvas from a prior fx_get_timeseries call. Required first step before fx_dataframe_query — use it to discover table names and column schemas. Requires DataCanvas, which is on by default — a deployment that turns it off does not list this tool at all, and fx_get_timeseries returns every range inline there.
- `fx_dataframe_query`: Run a read-only SQL SELECT against DataCanvas tables staged by fx_get_timeseries. Supports aggregations, GROUP BY, window functions, and JOINs across multiple registered tables. Run fx_dataframe_describe first to discover table names and column schemas. Requires DataCanvas, which is on by default — a deployment that turns it off does not list this tool at all, and fx_get_timeseries returns every range inline there.
- `fx_get_rate`: Get the exchange rate for a currency pair on a given date (default: latest). Returns the rate, the actual rate date (which may differ from the requested date on weekends/holidays — ECB publishes business days only), and source provenance. Cross-rates are triangulated through EUR automatically. A same-currency pair returns a rate of 1, dated to the same publication day any other pair would report for that date. Use fx_convert_currency when you want the converted amount; use this tool when you only need the rate number.
- `fx_get_rates`: Get all available exchange rates for one base currency in a single snapshot. Useful for bulk comparison and seeding downstream tools. Returns a map of quote currency → rate, the actual snapshot date, and whether a historical date was snapped from a weekend/holiday to the prior business day. Optionally filter to a subset of quote currencies via symbols. Listing the base currency itself in symbols is accepted and returns a rate of 1 for it.
- `fx_get_timeseries`: Get historical daily exchange rates for a currency pair over a date range. ECB publishes on business days only — weekends and holidays produce no entry, and no date outside the requested range is ever returned, so a range covering only non-publication days comes back with an empty rates map and a notice explaining why. A same-currency pair returns a rate of 1 on each publication day in the range. Inline results are returned as a date→rate map paged at 500 publication days: rate_count is always the total for the requested range, and when a page is cut short the response carries truncated=true and next_start_date — call again with start_date set to next_start_date and the same end_date for the next page. Long ranges (>90 days by default) are considered for DataCanvas staging: when the series exceeds the 40,000-character preview budget, it is staged and the response carries spilled=true, a canvas_id, and a table_name — call fx_dataframe_describe to inspect the staged table, then fx_dataframe_query to run SQL against it. On a deployment with DataCanvas turned off, long ranges are paged inline (spilled=false) and the notice says so.
- `fx_list_currencies`: List all supported ISO 4217 currency codes with their full names. Call this before converting to disambiguate "dollars" (USD vs AUD vs CAD vs HKD vs SGD) or to validate a user-supplied currency code. Covers the ~30 ECB reference currencies.