Phở Research Tools
Checks AI-written references against Crossref, PubMed and OpenAlex. Formats citations, PRISMA.
我该使用它吗
质量与安全性
A
基于对工具定义和协议合规性的自动分析。
上下文开销
~3,809token 数（工具定义）
~7.6 KB典型响应大小
对注意力有显著影响（占 128k 上下文窗口的 2.98%）
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": {
"pho-research-tools": {
"url": "https://mcp.pho.chat/mcp"
}
}
}远程端点
https://mcp.pho.chat/mcpstreamable-http它能做什么
工具清单
工具（5）
🟢 只读🟡 写入🔴 删除⚪ 未知
🟢verify_references(references)
Check every reference in a list against Crossref, PubMed and OpenAlex and report, for each one, whether the paper really exists and whether the details are right. Call this before presenting any reference list to the user, including references you wrote yourself. Input: a pasted reference list in any common style (APA, Vancouver, numbered, mixed), BibTeX, RIS, CSL-JSON, or one DOI per line. Each reference gets exactly one status: verified, verified_with_corrections (real paper, some details wrong; corrected metadata included), likely_fabricated (no such paper exists), not_found (could not be confirmed either way), retracted, or ambiguous (several papers fit). Never present a likely_fabricated, not_found or ambiguous reference as real, and always tell the user about a retracted one. At most 50 references per call.
输入模式
{
"type": "object",
"properties": {
"references": {
"type": "string",
"minLength": 1,
"maxLength": 100000,
"description": "The reference list exactly as written: any citation style, BibTeX, RIS, CSL-JSON, or DOIs one per line."
}
},
"required": [
"references"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}输出模式
{
"type": "object",
"properties": {
"summary": {
"type": "object",
"properties": {
"total": {
"type": "integer"
},
"verified": {
"type": "integer"
},
"verified_with_corrections": {
"type": "integer"
},
"likely_fabricated": {
"type": "integer"
},
"not_found": {
"type": "integer"
},
"retracted": {
"type": "integer"
},
"ambiguous": {
"type": "integer"
},
"not_checked": {
"type": "integer"
}
},
"required": [
"total",
"verified",
"verified_with_corrections",
"likely_fabricated",
"not_found",
"retracted",
"ambiguous",
"not_checked"
],
"additionalProperties": false
},
"references": {
"type": "array",
"items": {
"type": "object",
"properties": {
"index": {
"type": "integer"
},
"status": {
"type": "string",
"enum": [
"verified",
"verified_with_corrections",
"likely_fabricated",
"not_found",
"retracted",
"ambiguous",
"not_checked"
]
},
"confidence": {
"type": "number"
},
"summary": {
"type": "string"
},
"cited": {
"type": "string"
},
"corrected": {
"type": "object",
"properties": {
"title": {
"type": "string"
},
"authors": {
"type": "array",
"items": {
"type": "string"
}
},
"year": {
"type": "integer"
},
"journal": {
"type": "string"
},
"volume": {
"type": "string"
},
"issue": {
"type": "string"
},
"pages": {
"type": "string"
},
"doi": {
"type": "string"
},
"pmid": {
"type": "string"
}
},
"required": [
"title",
"authors"
],
"additionalProperties": false
},
"mismatches": {
"type": "array",
"items": {
"type": "object",
"properties": {
"field": {
"type": "string"
},
"cited": {
"type": "string"
},
"found": {
"type": "string"
}
},
"required": [
"field",
"cited",
"found"
],
"additionalProperties": false
}
},
"links": {
"type": "object",
"properties": {
"doi": {
"type": "string"
},
"pubmed": {
"type": "string"
},
"openalex": {
"type": "string"
}
},
"additionalProperties": true
},
"retraction": {
"type": "object",
"properties": {
"status": {
"type": "string"
},
"notices": {
"type": "array",
"items": {
"type": "object",
"properties": {
"type": {
"type": "string"
},
"label": {
"type": "string"
},
"date": {
"type": "string"
},
"source": {
"type": "string"
},
"notice_doi": {
"type": "string"
}
},
"required": [
"type",
"label",
"source"
],
"additionalProperties": false
}
}
},
"required": [
"status",
"notices"
],
"additionalProperties": false
}
},
"required": [
"index",
"status",
"summary",
"cited",
"mismatches",
"links"
],
"additionalProperties": false
}
},
"quota": {
"type": "object",
"properties": {
"plan": {
"type": "string",
"enum": [
"anon",
"free"
]
},
"metric": {
"type": "string",
"enum": [
"verifications",
"calls"
]
},
"used": {
"type": "integer"
},
"limit": {
"type": "integer"
},
"resets": {
"type": "string"
}
},
"required": [
"plan",
"metric",
"used",
"limit",
"resets"
],
"additionalProperties": false
}
},
"required": [
"summary",
"references",
"quota"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢format_citations(references, style, output, verify)
Format references in a citation style (APA 7, MLA 9, Chicago author-date, Harvard, Vancouver, IEEE, AMA) or export them as BibTeX, RIS or CSL-JSON. By default every reference is first verified against Crossref, PubMed and OpenAlex, and verified references are formatted from the database record, so wrong years, pages or author names in the input are corrected. References that could not be verified are formatted exactly as written and flagged in `entries[].warnings`; tell the user about them. Returns the bibliography and, for text and html output, the matching in-text citation for each reference.
输入模式
{
"type": "object",
"properties": {
"references": {
"type": "string",
"minLength": 1,
"maxLength": 100000,
"description": "References in any form: a pasted list in any style, BibTeX, RIS, CSL-JSON, or DOIs one per line."
},
"style": {
"type": "string",
"enum": [
"apa",
"mla",
"chicago",
"harvard",
"vancouver",
"ieee",
"ama"
],
"default": "apa",
"description": "Citation style: apa (APA 7), mla (MLA 9), chicago (Chicago author-date), harvard (Harvard), vancouver (Vancouver), ieee (IEEE), ama (AMA). Ignored for bibtex/ris/csl-json output."
},
"output": {
"type": "string",
"enum": [
"text",
"html",
"bibtex",
"ris",
"csl-json"
],
"default": "text",
"description": "text (plain bibliography), html, bibtex, ris or csl-json."
},
"verify": {
"type": "boolean",
"default": true,
"description": "Verify each reference before formatting (recommended). Set false only for references already verified in this conversation; they are then formatted exactly as given."
}
},
"required": [
"references"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}输出模式
{
"type": "object",
"properties": {
"style": {
"type": "string"
},
"style_title": {
"type": "string"
},
"output": {
"type": "string"
},
"content": {
"type": "string"
},
"entries": {
"type": "array",
"items": {
"type": "object",
"properties": {
"index": {
"type": "integer"
},
"verified": {
"type": "boolean"
},
"status": {
"type": "string"
},
"retracted": {
"type": "boolean"
},
"bibliography": {
"type": "string"
},
"in_text": {
"type": "string"
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
}
},
"required": [
"index",
"verified",
"retracted",
"bibliography",
"in_text",
"warnings"
],
"additionalProperties": false
}
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
},
"quota": {
"type": "object",
"properties": {
"plan": {
"type": "string",
"enum": [
"anon",
"free"
]
},
"metric": {
"type": "string",
"enum": [
"verifications",
"calls"
]
},
"used": {
"type": "integer"
},
"limit": {
"type": "integer"
},
"resets": {
"type": "string"
}
},
"required": [
"plan",
"metric",
"used",
"limit",
"resets"
],
"additionalProperties": false
}
},
"required": [
"style",
"style_title",
"output",
"content",
"entries",
"warnings"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢search_papers(query, year_from, year_to, open_access_only, study_types, ...)
Search PubMed and OpenAlex for real papers. Every result carries a DOI or PubMed ID that resolves, so nothing returned can be fabricated. Use it to find sources for a claim or a topic instead of recalling references from memory. Filters: publication year range, open-access only, study types, and the number of results. Retracted papers are left out by default. Preprints are included and marked `preprint: true`; tell the user a preprint is not peer reviewed. Results are ranked by the databases' relevance, alternating between the two sources.
输入模式
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 2,
"maxLength": 500,
"description": "What to look for, in English keywords (e.g. \"sepsis early antibiotics mortality\")."
},
"year_from": {
"type": "integer",
"minimum": 1800,
"maximum": 2100,
"description": "Earliest publication year, inclusive."
},
"year_to": {
"type": "integer",
"minimum": 1800,
"maximum": 2100,
"description": "Latest publication year, inclusive."
},
"open_access_only": {
"type": "boolean",
"default": false,
"description": "Only papers with a free, legal full text."
},
"study_types": {
"type": "array",
"items": {
"type": "string",
"maxLength": 60
},
"maxItems": 5,
"description": "Study designs. Known ids use database filters: rct, clinical-trial, systematic-review, meta-analysis, review, guideline, cohort, case-control, case-report, observational. Any other text is added as keywords."
},
"max_results": {
"type": "integer",
"minimum": 1,
"maximum": 50,
"default": 10,
"description": "Number of results, 1–50."
},
"include_retracted": {
"type": "boolean",
"default": false,
"description": "Keep retracted papers (flagged) instead of leaving them out."
}
},
"required": [
"query"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}输出模式
{
"type": "object",
"properties": {
"query": {
"type": "string"
},
"status": {
"type": "string",
"enum": [
"ok",
"unavailable"
]
},
"results": {
"type": "array",
"items": {
"type": "object",
"properties": {
"rank": {
"type": "integer"
},
"title": {
"type": "string"
},
"authors": {
"type": "array",
"items": {
"type": "string"
}
},
"year": {
"type": "integer"
},
"journal": {
"type": "string"
},
"doi": {
"type": "string"
},
"pmid": {
"type": "string"
},
"url": {
"type": "string"
},
"links": {
"type": "object",
"properties": {},
"additionalProperties": true
},
"open_access": {
"type": "object",
"properties": {
"is_oa": {
"type": "boolean"
},
"url": {
"type": "string"
}
},
"required": [
"is_oa"
],
"additionalProperties": false
},
"found_in": {
"type": "array",
"items": {
"type": "string"
}
},
"study_type_match": {
"type": "string"
},
"retracted": {
"type": "boolean"
},
"preprint": {
"type": "boolean"
},
"cited_by_count": {
"type": "integer"
}
},
"required": [
"rank",
"title",
"authors",
"url",
"links",
"open_access",
"found_in",
"retracted",
"preprint"
],
"additionalProperties": false
}
},
"excluded": {
"type": "object",
"properties": {
"no_identifier": {
"type": "integer"
},
"retracted": {
"type": "integer"
},
"notices": {
"type": "integer"
},
"outside_years": {
"type": "integer"
}
},
"required": [
"no_identifier",
"retracted",
"notices",
"outside_years"
],
"additionalProperties": false
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
}
},
"required": [
"query",
"status",
"results",
"excluded",
"warnings"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢check_retractions(dois)
Check DOIs for retractions, expressions of concern and corrections, using Crossref (which includes Retraction Watch data). Status per DOI: retracted, concern, corrected, other_notice, no_notice (nothing on record), not_in_crossref, invalid_doi, rate_limited or lookup_failed (could not be checked; say so, do not report it as clean). Use it before citing a paper whose status you are unsure of, or to screen a whole reference list. verify_references already includes this check.
输入模式
{
"type": "object",
"properties": {
"dois": {
"type": "array",
"items": {
"type": "string",
"minLength": 1,
"maxLength": 300
},
"minItems": 1,
"maxItems": 50,
"description": "DOIs, bare (10.1000/xyz) or as https://doi.org/ links."
}
},
"required": [
"dois"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}输出模式
{
"type": "object",
"properties": {
"results": {
"type": "array",
"items": {
"type": "object",
"properties": {
"doi": {
"type": "string"
},
"status": {
"type": "string"
},
"title": {
"type": "string"
},
"notices": {
"type": "array",
"items": {
"type": "object",
"properties": {
"type": {
"type": "string"
},
"label": {
"type": "string"
},
"date": {
"type": "string"
},
"source": {
"type": "string"
},
"notice_doi": {
"type": "string"
}
},
"required": [
"type",
"label",
"source"
],
"additionalProperties": false
}
}
},
"required": [
"doi",
"status",
"notices"
],
"additionalProperties": false
}
},
"counts": {
"type": "object",
"additionalProperties": {
"type": "integer"
}
},
"quota": {
"type": "object",
"properties": {
"plan": {
"type": "string",
"enum": [
"anon",
"free"
]
},
"metric": {
"type": "string",
"enum": [
"verifications",
"calls"
]
},
"used": {
"type": "integer"
},
"limit": {
"type": "integer"
},
"resets": {
"type": "string"
}
},
"required": [
"plan",
"metric",
"used",
"limit",
"resets"
],
"additionalProperties": false
}
},
"required": [
"results",
"counts",
"quota"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢prisma_flow_diagram(counts, other_methods_column, footnotes)
Draw a PRISMA 2020 flow diagram for a new systematic review from the record counts, as a PNG image plus an SVG and download links. Give the counts you have; boxes that follow arithmetically from others are filled in and marked as derived. The tool checks that the numbers add up (records identified minus removed equals screened, and so on) and returns a warning naming each box that does not, with the value it should have. Show those warnings to the user; the diagram always shows the numbers exactly as given. Add `otherMethods` counts to draw the second column (websites, organisations, citation searching).
输入模式
{
"type": "object",
"properties": {
"counts": {
"type": "object",
"properties": {
"databases": {
"type": "integer",
"minimum": 0,
"maximum": 100000000,
"description": "Records identified from databases (total)."
},
"databaseSources": {
"type": "array",
"items": {
"type": "object",
"properties": {
"name": {
"type": "string",
"minLength": 1,
"maxLength": 160
},
"n": {
"$ref": "#/properties/counts/properties/databases"
}
},
"required": [
"name",
"n"
],
"additionalProperties": false
},
"maxItems": 30,
"description": "Per-database counts, e.g. [{\"name\":\"MEDLINE\",\"n\":812}]."
},
"registers": {
"$ref": "#/properties/counts/properties/databases",
"description": "Records identified from registers."
},
"registerSources": {
"type": "array",
"items": {
"$ref": "#/properties/counts/properties/databaseSources/items"
},
"maxItems": 30
},
"duplicatesRemoved": {
"$ref": "#/properties/counts/properties/databases"
},
"automationIneligible": {
"$ref": "#/properties/counts/properties/databases",
"description": "Records marked ineligible by automation tools."
},
"otherRemoved": {
"$ref": "#/properties/counts/properties/databases",
"description": "Records removed for other reasons before screening."
},
"recordsScreened": {
"$ref": "#/properties/counts/properties/databases"
},
"recordsExcluded": {
"$ref": "#/properties/counts/properties/databases"
},
"reportsSought": {
"$ref": "#/properties/counts/properties/databases",
"description": "Reports sought for retrieval."
},
"reportsNotRetrieved": {
"$ref": "#/properties/counts/properties/databases"
},
"reportsAssessed": {
"$ref": "#/properties/counts/properties/databases",
"description": "Reports assessed for eligibility."
},
"reportsExcludedReasons": {
"type": "array",
"items": {
"$ref": "#/properties/counts/properties/databaseSources/items"
},
"maxItems": 30,
"description": "Reports excluded, one entry per reason."
},
"reportsExcluded": {
"$ref": "#/properties/counts/properties/databases",
"description": "Total reports excluded, when reasons are not itemised."
},
"studiesIncluded": {
"$ref": "#/properties/counts/properties/databases",
"description": "Studies included in the review."
},
"reportsIncluded": {
"$ref": "#/properties/counts/properties/databases",
"description": "Reports of included studies."
},
"otherMethods": {
"type": "object",
"properties": {
"websites": {
"$ref": "#/properties/counts/properties/databases"
},
"organisations": {
"$ref": "#/properties/counts/properties/databases"
},
"citationSearching": {
"$ref": "#/properties/counts/properties/databases"
},
"sources": {
"type": "array",
"items": {
"$ref": "#/properties/counts/properties/databaseSources/items"
},
"maxItems": 30
},
"reportsSought": {
"$ref": "#/properties/counts/properties/databases"
},
"reportsNotRetrieved": {
"$ref": "#/properties/counts/properties/databases"
},
"reportsAssessed": {
"$ref": "#/properties/counts/properties/databases"
},
"reportsExcludedReasons": {
"type": "array",
"items": {
"$ref": "#/properties/counts/properties/databaseSources/items"
},
"maxItems": 30
},
"reportsExcluded": {
"$ref": "#/properties/counts/properties/databases"
}
},
"additionalProperties": false,
"description": "Counts for the \"identification via other methods\" column."
}
},
"additionalProperties": false
},
"other_methods_column": {
"type": "boolean",
"description": "Force the other-methods column on or off. Default: drawn when otherMethods counts are given."
},
"footnotes": {
"type": "boolean",
"default": true,
"description": "Print the template footnotes and source attribution."
}
},
"required": [
"counts"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}输出模式
{
"type": "object",
"properties": {
"image_url": {
"type": "string"
},
"svg_url": {
"type": "string"
},
"svg": {
"type": "string"
},
"width": {
"type": "number"
},
"height": {
"type": "number"
},
"warnings": {
"type": "array",
"items": {
"type": "object",
"properties": {
"code": {
"type": "string"
},
"field": {
"type": "string"
},
"message": {
"type": "string"
},
"expected": {
"type": "number"
},
"given": {
"type": "number"
}
},
"required": [
"code",
"field",
"message"
],
"additionalProperties": false
}
},
"counts": {
"type": "object",
"additionalProperties": {
"type": "object",
"properties": {
"value": {
"type": "number"
},
"derived": {
"type": "boolean"
}
},
"required": [
"derived"
],
"additionalProperties": false
}
}
},
"required": [
"svg",
"width",
"height",
"warnings",
"counts"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}社区
证据
最近观测
已验证未记录版本5 个工具