carbon-fyi
Voluntary carbon market news and open carbon price data. Read-only, keyless.
我该使用它吗
质量与安全性
基于对工具定义和协议合规性的自动分析。
上下文开销
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高,可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中:
{
"mcpServers": {
"carbon-fyi": {
"url": "https://www.carbon.fyi/api/mcp"
}
}
}远程端点
https://www.carbon.fyi/api/mcpstreamable-http它能做什么
工具清单
工具(6)
🟢search_news(query, tag, since, limit)
Search carbon.fyi's curated voluntary carbon market story archive. Each story is an editorially clustered summary of one development, written from one or more original reports, and always carries the original source name and URL. Supply at least one filter; results are newest-first. Use `get_market_summary` first if you do not know what tags or date range to ask for.
输入模式
{
"type": "object",
"properties": {
"query": {
"description": "case-insensitive text match against headline and summary, e.g. 'Verra' or 'biochar'",
"type": "string",
"minLength": 2,
"maxLength": 120
},
"tag": {
"description": "restrict to one story category",
"type": "string",
"enum": [
"registry-action",
"methodology",
"corporate-deal",
"integrity",
"article-6",
"market-data",
"policy-regulatory",
"other-vcm"
]
},
"since": {
"description": "only stories first published on or after this UTC day",
"type": "string",
"pattern": "^\\d{4}-\\d{2}-\\d{2}$"
},
"limit": {
"description": "max stories to return (default 20, ceiling 50)",
"type": "integer",
"minimum": 1,
"maximum": 50
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_story(id)
Fetch a single story by its id, including every source that reported it. Story ids are permanent — a published story keeps its id and its public URL indefinitely — so they are safe to store.
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"minLength": 1,
"maxLength": 200,
"description": "the story's `id`, as returned by search_news"
}
},
"required": [
"id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_carbon_prices
Current voluntary carbon credit prices from two independent sources: per-category minimum/median price and available supply across the Carbonmark marketplace catalog, and spot prices per Klima Protocol carbon class. The two are NOT comparable — read the `caveats` field on each before using either number.
输入模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_price_history(kind, key, days)
Daily-averaged historical price series. This archive exists only because carbon.fyi records it three times a day: no upstream API serves a historical carbon price series, so these observations cannot be obtained anywhere else and cannot be re-derived. Check `seriesStartsAt` in the response — the archive began on 2026-08-27, so early windows are necessarily short.
输入模式
{
"type": "object",
"properties": {
"kind": {
"type": "string",
"enum": [
"carbonmark-category",
"klima-class"
],
"description": "which price source: marketplace category medians, or Klima carbon-class spot prices"
},
"key": {
"description": "restrict to one category or class name (exact match, as returned by get_carbon_prices); omit for all",
"type": "string",
"maxLength": 120
},
"days": {
"description": "look-back window in days (default 90)",
"type": "integer",
"minimum": 1,
"maximum": 730
}
},
"required": [
"kind"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_market_summary
One call for orientation: how much of the archive is available, when it last updated, the story-tag breakdown (use these as `tag` values for search_news), and the current price table. Start here.
输入模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢list_sources
The publications, registries and integrity bodies carbon.fyi monitors. Useful for judging coverage — a topic absent from the archive may simply not be covered by these sources.
输入模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}社区
证据