CourierPR
Search and read published company press releases across six newsroom desks and twenty categories.
我該用這個嗎
品質與安全性
發現項目(1)
- LOW在 fetch 中
根據工具定義與協定合規性的自動化分析。
上下文成本
這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高,可用於其他工作的注意力就越少。
安裝
一鍵安裝
將以下內容加入你的 `claude_desktop_config.json` 檔案:
{
"mcpServers": {
"mcp": {
"url": "https://courierpr.com/api/mcp"
}
}
}遠端端點
https://courierpr.com/api/mcpstreamable-http它能做什麼
工具清單
工具(7)
🟢list_desks
The six newsroom desks and the twenty categories under them, with slugs and links. Use it to resolve a topic the user named ('tech', 'health') into a desk or category before calling latest_releases. Returns no releases.
輸入結構描述
{
"type": "object",
"properties": {},
"required": [],
"additionalProperties": false
}🟢latest_releases(desk, category, limit)
The newest published press releases, optionally within one desk or one category. Use it for 'what is new in tech', 'latest funding announcements', 'recent releases on CourierPR'. Each result is a title, company, date, short summary and a link; the full text is on the linked page. Limitation: at most 10 per call, newest first, no date filter.
輸入結構描述
{
"type": "object",
"properties": {
"desk": {
"type": "string",
"description": "Desk slug or name from list_desks, e.g. science-tech"
},
"category": {
"type": "string",
"description": "Category slug or name from list_desks, e.g. business-finance. Takes precedence over desk."
},
"limit": {
"type": "integer",
"description": "How many to return",
"minimum": 1,
"maximum": 10,
"default": 8
}
},
"required": [],
"additionalProperties": false
}🟢search_releases(query, category, limit)
Full-text search over every published press release (title and body), optionally within one category. Use it when the user names a company, product, person or topic. Returns short summaries with links; the full text is on the linked page. Limitation: English full-text matching, at most 10 results, no date filter.
輸入結構描述
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Words to search for; quotes and minus work like a search engine"
},
"category": {
"type": "string",
"description": "Optional category slug or name from list_desks"
},
"limit": {
"type": "integer",
"description": "How many to return",
"minimum": 1,
"maximum": 10,
"default": 8
}
},
"required": [
"query"
],
"additionalProperties": false
}🟢get_release(slug)
One published press release by slug (the last part of its URL) or by its full courierpr.com URL: title, company, category, dateline, date, the opening 120 words and a link to the full release. Use it after a search or list result when the user wants more on one item. Limitation: returns a summary, not the full text; media contact details are never included and are on the page behind a human check.
輸入結構描述
{
"type": "object",
"properties": {
"slug": {
"type": "string",
"description": "Release slug, e.g. acme-raises-series-b, or the full release URL"
}
},
"required": [
"slug"
],
"additionalProperties": false
}🟢how_to_file
What filing a release on CourierPR involves: what is accepted (the press code) and the links to file, for agencies and for journalists. Use it when the user asks how to publish or distribute a press release. Returns links only; it does not file anything.
輸入結構描述
{
"type": "object",
"properties": {},
"required": [],
"additionalProperties": false
}🟢search(query)
Connector-style search over published press releases. Returns ids for fetch. Prefer search_releases when you can filter by category.
輸入結構描述
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Words to search for"
}
},
"required": [
"query"
],
"additionalProperties": false
}🟢fetch(id)
Connector-style fetch of one press release by the id from search. Same summary as get_release.
輸入結構描述
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Release slug from search"
}
},
"required": [
"id"
],
"additionalProperties": false
}社群
證據