# io.github.cyanheads/orcid-mcp-server

> Researcher profiles, works, affiliations, funding, and peer reviews from the ORCID registry.

- name: io.github.cyanheads/orcid-mcp-server
- version: 0.3.3
- connection class: R0 (autonomous)
- trust: 94/100
- quality: 70/100 (needs work)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/io.github.cyanheads%2Forcid-mcp-server
- tools: https://api.protogrid.dev/v1/servers/io.github.cyanheads%2Forcid-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": {
    "orcid-mcp-server": {
      "type": "http",
      "url": "https://orcid.caseyjhand.com/mcp"
    }
  }
}
```


## Trust

- hygiene: 100
- liveness: 100
- freshness: 100
- provenance: 80
- drivers: +repository +namespace-matches-repo +reachable +uptime-100% +updated-2d-ago
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 70/100 (needs work)

### 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`: 2 tools have descriptions over 1,024 characters, which crowds the context
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 9 of 9 tools declare readOnlyHint or destructiveHint
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: orcid_get_profile (destructiveHint); orcid_search_researchers (destructiveHint); orcid_get_funding (destructiveHint); …
- pass `tools.token_cost`: about 4,526 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 25 (weight 15)

- warn `stability.changes`: 27 tool changes in 30 days
- fail `stability.rug_pull`: 2 tools kept their name but changed most of their description in 30 days; review before trusting them

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

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: deps.dev does not know @cyanheads/orcid-mcp-server@0.3.3, @cyanheads/orcid-mcp-server@0.3.3

- tools: 9, about 4,526 tokens to load them all
- tool set hash: fdb028d283313b05f4c4b94e111b4536
- tools last changed: 2026-10-19T02:38:18.058Z
- badge: [![protogrid quality](https://protogrid.dev/badge/io.github.cyanheads/orcid-mcp-server.svg)](https://protogrid.dev/servers/io.github.cyanheads/orcid-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

- 2026-10-19 `orcid_resolve_researcher` changed (inputSchema)
- 2026-10-19 `orcid_get_research_resources` changed (description, inputSchema, outputSchema) **meaning changed**
  - before: List research resources associated with an ORCID researcher — compute allocations, equipment access, lab facilities, data resources, and clinical study registrations. This is a newer ORCID section; most researchers have no entries. Returns the resource title, hosting organization, external identifiers (often a URI to the allocation portal), access period, and the source that added each entry. Most entries are deposited by resource-allocation systems (e.g. ACCESS, XSEDE) rather than researchers themselves.
  - after: List allocation proposals for an ORCID researcher, or supply put_code from a summary record to retrieve its actual resource items. Summary is the default view and includes every proposal host, identifiers, access period, and source provenance; hostOrganization remains the first proposal host. Detail adds item names, types, all operators, identifiers, and URLs. Page items with item_offset (default 0) and item_limit (default 50, effective maximum 100, hard maximum 10,000), both requiring put_code; continue at nextItemOffset with the same selector. Both detail surfaces fit 64,000 bytes including metadata and continuation, with an explicit budgetOverflow exception for an indivisible oversized item returned alone or oversized proposal metadata. Most researchers have no entries; allocation systems such as ACCESS deposit these records.
- 2026-10-19 `orcid_get_peer_reviews` changed (inputSchema, outputSchema)
- 2026-10-19 `orcid_get_affiliations` changed (outputSchema)
- 2026-10-19 `orcid_get_work_detail` changed (description, inputSchema, outputSchema) **meaning changed**
  - before: Fetch full detail records for 1–100 works by their put-codes in a single request. Put-codes are returned by orcid_get_works in the putCode field of each work entry. Returns the abstract (short-description), contributors with CRediT roles, the external IDs on that record (DOI, PMID, arXiv, ISBN, etc.), citation metadata (BibTeX or other formats when provided), journal title, URL, and the source that added it for each work. A record listing more than 100 contributors keeps the first 100 plus the requested researcher's own entries, with contributorCount giving the total and contributorsTruncated set; a citation over 8,192 bytes is left out and flagged by citationOmitted. Per-record errors (not-found or inaccessible put-codes) are surfaced as error entries rather than failing the whole call. Records are added while the response stays within its 64,000-byte budget; put-codes left out are returned in deferredPutCodes to pass back in a follow-up call. A repeated put-code is fetched once.
  - after: Fetch full work details by put-code from orcid_get_works. Accepts 1–1,000 raw codes, deduplicates in first-occurrence order, and fetches the first 100 distinct codes in one request. Returns abstracts, contributors with CRediT roles, the record's external IDs (DOI, PMID, arXiv, ISBN, etc.), citation metadata, journal title, URL, and source provenance. A record listing more than 100 contributors keeps the first 100 plus the requested researcher's own entries, with contributorCount and contributorsTruncated; a citation over 8,192 bytes is left out and flagged by citationOmitted. Per-record errors are returned alongside successful works. Records are admitted under a 64,000-byte budget on each complete response surface. Every requested distinct code unsettled by an admitted work or identified error is listed in deferredPutCodes in request order, including input-cap, byte-budget, and absent-upstream-result deferrals; pass those codes in another call. Notices distinguish the causes; no automatic chunk fetching occurs.
- 2026-10-19 `orcid_get_works` changed (inputSchema, outputSchema)
- 2026-10-19 `orcid_search_researchers` changed (inputSchema, outputSchema)
- 2026-10-03 `orcid_resolve_researcher` changed (description, inputSchema, outputSchema)
  - before: Disambiguate an author name to a verified ORCID iD. Returns ranked candidates (5 by default, up to 20 via the rows parameter) with transparent disambiguation signals: name match type (exact/partial/other-name/none), institution overlap flag, and whether a DOI or PMID anchor was used in the query. A DOI or PMID anchor is near-deterministic — it filters to researchers who have linked that specific work to their ORCID record. Use this tool (not orcid_search_researchers) when the input is an ambiguous name that needs ranked disambiguation. No synthetic scores are used — raw signals only.
  - after: Disambiguate an author name to a verified ORCID iD. Returns ranked candidates (5 by default, up to 20 via the rows parameter) with transparent disambiguation signals: name match type (exact/partial/other-name/none), institution overlap flag, and whether a DOI or PMID anchor was used in the query. The name is searched as an exact phrase first; when that finds nothing, byline initials (J. Doudna, Jennifer A. Doudna) and then the other names listed on ORCID records are tried. A DOI or PMID anchor is near-deterministic — it filters to researchers who have linked that specific work to their ORCID record. Use this tool (not orcid_search_researchers) when the input is an ambiguous name that needs ranked disambiguation. No synthetic scores are used — raw signals only.
- 2026-10-03 `orcid_get_research_resources` changed (description, inputSchema, outputSchema)
  - before: List research resources associated with an ORCID researcher — compute allocations, equipment access, lab facilities, data resources, and clinical study registrations. This is a newer ORCID section; most researchers have no entries. Returns the resource title, hosting organization, external identifiers (often a URI to the allocation portal), and access period. Most entries are deposited by resource-allocation systems (e.g. ACCESS, XSEDE) rather than researchers themselves.
  - after: List research resources associated with an ORCID researcher — compute allocations, equipment access, lab facilities, data resources, and clinical study registrations. This is a newer ORCID section; most researchers have no entries. Returns the resource title, hosting organization, external identifiers (often a URI to the allocation portal), access period, and the source that added each entry. Most entries are deposited by resource-allocation systems (e.g. ACCESS, XSEDE) rather than researchers themselves.
- 2026-10-03 `orcid_get_peer_reviews` changed (description, inputSchema, outputSchema)
  - before: Fetch peer review activity for an ORCID researcher: convening organizations (journals and publishers), reviewer role (reviewer, editor, chair, etc.), review type, completion dates, and ISSN-keyed group identifiers. Use to assess editorial activity, journal affiliations, and the scope of a researcher's peer review contributions. Peer review records are self-reported or imported by participating publishers — coverage varies by researcher.
  - after: Fetch peer review activity for an ORCID researcher: reviewer role (reviewer, editor, chair, etc.), review type, completion dates, ISSN-keyed group identifiers, and the convening organization as ORCID records it — often the service that imported the review (e.g. Publons, Clarivate). ORCID returns no journal name; groupIssn is the key to the journal, and a review that has one is headed by it. Use to assess editorial activity, journal affiliations, and the scope of a researcher's peer review contributions. Reviews come from the researcher or from member organizations that import them; sources names who added each, and coverage varies by researcher.

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

## Dependencies

- npm @cyanheads/orcid-mcp-server@0.3.3: unknown
- npm @cyanheads/orcid-mcp-server@0.3.3: 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://orcid.caseyjhand.com/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

- npm @cyanheads/orcid-mcp-server@0.3.3
- npm @cyanheads/orcid-mcp-server@0.3.3

## Tools (9)

- `orcid_get_affiliations`: Fetch affiliation records for an ORCID researcher. The `types` parameter controls which affiliation sections to return: employment, education, invited-positions, distinctions, memberships, qualifications, services, or all. Default is employment and education. Returns organization names, disambiguated organization identifiers (ROR/GRID/Ringgold), departments, roles, date ranges, and sources. Affiliations come from the researcher or from member organizations such as a university research information system; sources names who added each, and selfAsserted tells the two apart. Absence does not mean no affiliation.
- `orcid_get_funding`: Fetch funding records for an ORCID researcher: grants, contracts, awards, and salary awards. Returns funder names, funder organization identifiers, grant numbers, funding periods, and sources. Funding items come from the researcher or from member organizations such as funders, institutions, and search-and-link tools; sources names who added each. Most researchers have no funding entries even when they hold grants, so absence of funding records does not imply absence of funding. When records exist they are high-value for grant tracking and funder analysis.
- `orcid_get_peer_reviews`: Fetch peer review activity for an ORCID researcher: reviewer role (reviewer, editor, chair, etc.), review type, completion dates, ISSN-keyed group identifiers, and the convening organization as ORCID records it — often the service that imported the review (e.g. Publons, Clarivate). ORCID returns no journal name; groupIssn is the key to the journal, and a review that has one is headed by it. Use to assess editorial activity, journal affiliations, and the scope of a researcher's peer review contributions. Reviews come from the researcher or from member organizations that import them; sources names who added each, and coverage varies by researcher.
- `orcid_get_profile`: Fetch a researcher's public profile from ORCID: name, other names, biography, keywords, researcher URLs, and external identifiers such as Scopus Author ID, ResearcherID, and Loop profile. This is the entry point for building a researcher dossier. Pass a bare ORCID iD (0000-0001-2345-6789) or a full URI (https://orcid.org/0000-0001-2345-6789). The profile contains only publicly visible data — researchers control visibility per field.
- `orcid_get_research_resources`: List allocation proposals for an ORCID researcher, or supply put_code from a summary record to retrieve its actual resource items. Summary is the default view and includes every proposal host, identifiers, access period, and source provenance; hostOrganization remains the first proposal host. Detail adds item names, types, all operators, identifiers, and URLs. Page items with item_offset (default 0) and item_limit (default 50, effective maximum 100, hard maximum 10,000), both requiring put_code; continue at nextItemOffset with the same selector. Both detail surfaces fit 64,000 bytes including metadata and continuation, with an explicit budgetOverflow exception for an indivisible oversized item returned alone or oversized proposal metadata. Most researchers have no entries; allocation systems such as ACCESS deposit these records.
- `orcid_get_work_detail`: Fetch full work details by put-code from orcid_get_works. Accepts 1–1,000 raw codes, deduplicates in first-occurrence order, and fetches the first 100 distinct codes in one request. Returns abstracts, contributors with CRediT roles, the record's external IDs (DOI, PMID, arXiv, ISBN, etc.), citation metadata, journal title, URL, and source provenance. A record listing more than 100 contributors keeps the first 100 plus the requested researcher's own entries, with contributorCount and contributorsTruncated; a citation over 8,192 bytes is left out and flagged by citationOmitted. Per-record errors are returned alongside successful works. Records are admitted under a 64,000-byte budget on each complete response surface. Every requested distinct code unsettled by an admitted work or identified error is listed in deferredPutCodes in request order, including input-cap, byte-budget, and absent-upstream-result deferrals; pass those codes in another call. Notices distinguish the causes; no automatic chunk fetching occurs.
- `orcid_get_works`: Retrieve works associated with an ORCID iD — publications, datasets, software, preprints, and more. Returns work summaries with put-codes, titles, types, publication dates, journal names, and all external identifiers (DOIs, PMIDs, arXiv IDs, ISBNs). The first 50 works are returned by default; workCount reports the total available, and prolific records are paged with offset and the returned nextOffset (or raise limit). A page also stops early once the response reaches its 64,000-byte budget, so returnedCount can come in below limit; truncated and nextOffset then carry the continuation. Set include_external_ids to false to omit identifier lists for a lighter payload. Pass the putCode from each work to orcid_get_work_detail to retrieve the full record including abstract and contributors. External IDs are ready for chaining to Crossref, PubMed, or arXiv servers. Works come from the researcher or from member organizations such as Crossref or a university system; each work lists its sources and whether the researcher asserted it (selfAsserted). A researcher may not have linked all their publications.
- `orcid_resolve_researcher`: Disambiguate an author name to a verified ORCID iD. Returns ranked candidates (5 by default, up to 20 via the rows parameter) with transparent disambiguation signals: name match type (exact/partial/other-name/none), institution overlap flag, and whether a DOI or PMID anchor was used in the query. The name is searched as an exact phrase first; when that finds nothing, byline initials (J. Doudna, Jennifer A. Doudna) and then the other names listed on ORCID records are tried. A DOI or PMID anchor is near-deterministic — it filters to researchers who have linked that specific work to their ORCID record. Use this tool (not orcid_search_researchers) when the input is an ambiguous name that needs ranked disambiguation. No synthetic scores are used — raw signals only.
- `orcid_search_researchers`: Search the ORCID registry using structured field parameters or raw Solr syntax. Provide at least one non-blank search field; all provided structured params are ANDed together. The `query` field adds raw Solr syntax, ANDed with the structured params as one group. Returns ORCID iDs with inline name and institution data — no follow-up profile fetches needed for basic disambiguation. For ranked disambiguation of an ambiguous author name, use orcid_resolve_researcher instead. The ORCID Public API caps results at 10,000 — use pagination for large result sets. Each response also stops at 64,000 bytes; continue a cut page from nextStart.