Semantic Scholar for AI agents
Semantic Scholar is an academic search engine that indexes scientific papers, letting researchers find papers, authors, and citation relationships across fields of study. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Semantic Scholar.
Search for papers on a topic
Finds papers matching a research topic, a read-only lookup with no side effects on any record.
Get a paper's citations and references
Pulls the citing and cited papers for a specific paper to map its place in the literature.
Get recommendations for a paper
Suggests related papers based on a given paper, useful for expanding a literature review.
21 Semantic Scholar actions, graded by risk.
Every Semantic Scholar action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
11Look things up. Allowed by default.
- semantic_scholar.get_datasetGet dataset download links
- semantic_scholar.get_releaseGet dataset release information
- semantic_scholar.list_releasesList available dataset releases
- semantic_scholar.search_papersSearch papers by relevance
- semantic_scholar.get_dataset_diffsGet dataset diffs
- semantic_scholar.search_bulk_papersSearch Bulk Papers
- semantic_scholar.get_paper_recommendationsGet paper recommendations
- semantic_scholar.search_for_authors_by_nameSearch for authors by name
- semantic_scholar.get_recommendations_for_paperGet recommendations for paper
- semantic_scholar.get_details_for_multiple_papers_at_onceGet details for multiple papers at once
- semantic_scholar.get_details_for_multiple_authors_at_onceGet details for multiple authors at once
write
10Create and change things. Allow, or ask the user first.
- semantic_scholar.paper_title_searchPaper title search
- semantic_scholar.text_snippet_searchText snippet search
- semantic_scholar.details_about_a_paperDetails about a paper
- semantic_scholar.paper_relevance_searchPaper relevance search (Deprecated)
- semantic_scholar.details_about_an_authorDetails about an author
- semantic_scholar.details_about_a_paper_s_authorsDetails about a paper s authors
- semantic_scholar.suggest_paper_query_completionsSuggest paper query completions
- semantic_scholar.details_about_an_author_s_papersDetails about an author s papers
- semantic_scholar.details_about_a_paper_s_citationsDetails about a paper s citations
- semantic_scholar.details_about_a_paper_s_referencesDetails about a paper s references
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
Semantic Scholar in three steps.
- 01Your users connect Semantic ScholarThey add their Semantic Scholar api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “paper title search” can wait for the user to approve.
- 03Any agent can actYour agent calls Semantic Scholar through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('semantic_scholar', { read: 'allow', write: 'ask', // paper_title_search destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Semantic Scholar connects.
Users add their Semantic Scholar api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Semantic Scholar connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Semantic Scholar from any agent.
Semantic Scholar and Arc0, answered.
Can I use Semantic Scholar with Claude, ChatGPT or Cursor?
Yes. Connect Semantic Scholar to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Semantic Scholar actions you allow.
How do users connect Semantic Scholar?
Users add their Semantic Scholar api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Semantic Scholar actions can my agent take?
21 in total: 11 read, 10 write and 0 destructive, such as “paper title search”. Your policies decide which of them each agent may call.
Can I make my agent read-only in Semantic Scholar?
Yes. Allow read actions and deny writes in the Semantic Scholar policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call Semantic Scholar too?
Yes. The same Semantic Scholar connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Semantic Scholar into your agent.
Your users connect Semantic Scholar once, under your brand. Your agent gets 21 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan