# Search papers, filings or scholarly sources

> Search papers, filings or scholarly sources: 2 companies fully do it, 13 with a limit, 63 carry the tag. Every listing in Web search APIs with the reason and a source.

- Canonical: https://www.anchorterminal.com/niches/search.specialist-sources/
- Markdown: https://www.anchorterminal.com/niches/search.specialist-sources/index.md (~1,550 tokens)
- Slim: https://www.anchorterminal.com/niches/search.specialist-sources/index.min.md (~380 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/niches/search.specialist-sources/index.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-15

Two or three companies' docs show a full fit.

| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |
| --- | --- | --- | --- | --- |
| 2 | 13 | 63 | 5 | 0 |

## What counts as done

Results limited to the asked source type with the passage and the canonical link (DOI, filing URL), in one call.

How agents ask for it: find papers on X since 2024; what did the 10-K say about X; search clinical trials for X; find patents on X

## Does the job

The docs show it meets the test above, and nothing we read says otherwise.

- **TinyFish Search (Tiny Fish, Inc.)**: domain_type research_paper returns authors, venue, publication year, citation count and pdf_url, scoped by pub_year_min and pub_year_max, at no charge.
- **Valyu (Valyu.Network Ltd)**: included_sources takes dataset ids (valyu/valyu-pubmed, arXiv, SEC filings, trials, patents) or presets such as academic, with start_date and end_date and max_price capping the cost.

## Does the job, with a limit

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

- **Web Search Tool on Amazon Bedrock AgentCore (Amazon Web Services, Inc. (regional AWS entities by account location))**: filters.domainFilter.include (up to 100 domains) can confine results to arxiv.org or sec.gov, but there is no source-type mode and no DOI or filing metadata.
- **Exa API + MCP (Exa Labs Inc.)**: category publication keeps results to papers with author and venue metadata, and the data guide covers patents and trials, but no DOI or canonical link field was seen.
- **Liner Search MCP (Liner)**: Scholar Search returns papers with authors, journal, citationCount and date at $0.30 per 1,000, plus legal, finance, medical and business tools, but there is no free tier, the map's blocker for Liner.
- **Linkup (Linkup Technologies SAS)**: includeDomains can confine a search to arxiv.org or sec.gov, but there is no source-type mode or DOI field.
- **Mojeek Web Search API (Mojeek Limited)**: fi and fs include or exclude up to 25 domains; no source-type mode.
- **Parallel Search and Task APIs (Parallel Web Systems Inc.)**: include_domains is a hard allow list for a known publisher; no source-type mode or DOI metadata.
- **Perplexity Search API (Perplexity AI, Inc.)**: search_domain_filter takes up to 20 domains to include or exclude; no source-type mode or DOI field.
- **SerpApi (SerpApi, LLC)**: tbm=pts patents and bks books, and Google Scholar is a separate engine, but results are SERP entries without passages or DOIs.
- **Serper (Serper)**: Scholar and patents verticals are listed on the home page; their parameters and fields are undocumented.
- **SerpKite (Anastasia Klimova, trading as SerpKite)**: /v1/scholar and /v1/patents verticals exist; passages and DOI fields were not confirmed.
- **SpecStream (SpecStream SAS)**: Research lists cover papers, trials, grants and tenders as Deep Search rows, not a one-call search with DOIs.
- **Tavily API + MCP (AlphaAI Technologies Inc. dba Tavily)**: include_domains with include_domains_mode restrict limits results to listed sites; no source-type mode or DOI field.
- **You.com APIs (SuSea, Inc., d/b/a you.com)**: include_domains up to 500 and a Finance Research API, but no scholar or filings mode on search.

## Couldn't check

The docs were closed to us, or silent. These carry no mark either way.

- **DeepSeek Search and Fetch API (Hangzhou DeepSeek Artificial Intelligence Co., Ltd.)**: Docs not opened; no account.
- **SearchAPI.io (SearchApi, LLC)**: Only the Google engine OpenAPI was read; scholar or patents engines were not opened.

- not this job: Arcmira (Arcmira, Inc.): No papers, filings, trials or patents; the sources are shows and channels.
- not this job: Grounding with Bing Search (Microsoft Corporation): No source-type mode; Bing Custom Search restricts domains through a configured instance, not a request parameter.
- not this job: Brave Search API + MCP (Brave Software, Inc.): No scholar, filings or source-type mode; Goggles re-rank rather than restrict.
- not this job: Cloudflare Web Search API (Cloudflare, Inc.): No domain, date or source-type filters.
- not this job: Fetch (MCP reference server) (Model Context Protocol, a Series of LF Projects, LLC): No search.
- not this job: Grep AI (Parcha Labs, Inc.): No source-type search; built-in agents cover compliance research rather than papers or filings lookups.
- not this job: Kagi Search API (Kagi Inc.): Lenses restrict sites; no source-type mode or DOI metadata.
- not this job: NewsCatcher (NewsCatcher, Inc.): News publishers only.
- not this job: SERPHouse (Patoliya Infotech LLP): The Google verticals listed cover news, images, shopping, jobs, videos, forums and autocomplete; no scholar or patents.

## What stops agents finishing

- per-result credit price with no cap set (Valyu without max_price)
- no free tier (Liner)

## Around it

After: search.fetch-url, document-extraction.pdf-to-text.

## Where this comes from

Weight 5 of 100 in Web search APIs: No first-hand account found; weight rests on vendor pages for Valyu, Liner, TinyFish, Exa categories and Serper scholar, so treat as a guess

As data: https://www.anchorterminal.com/api/v1/use-cases/search.specialist-sources.json, or the MCP tool get_use_case with id search.specialist-sources. Draft, read 2026-11-12 from public docs and first-hand accounts, not yet reviewed by a person.

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp
