Sort

AgentEarth Tool Catalog

InformationRetrieval

356Tools
14tags
35Providers

Connect once to Google, Bing, Tavily, Exa, Firecrawl and more for web search, full-site crawling and structured extraction at lower cost.

  1. High-frequency retrieval and complex crawling are optimized for stability to reduce mid-task failures.
  2. Start search and crawling in one place for deep research, competitor monitoring, sentiment tracking and information collection.

Tool List

356 tools

ae_firecrawl_research_related_papers

Find arXiv papers through the citation graph and rank and filter candidates by research intent. Required parameters: `seed_ids` is an array of 1–10 actual paper IDs obtained from paper search or inspection; the first is the primary seed and later IDs are anchors. `intent` is a nonempty natural-language description of the research topic or relevance criteria. Optional parameters: `mode` is `similar` (default; co-citation/bibliographic coupling), `citers` (papers citing a seed), or `references` (papers cited by a seed). `k` is an integer from 1 to 500 specifying the maximum number returned. `rerank` is a boolean that enables an additional rerank over fused candidates. Defaults for `k` and `rerank` are not specified. Additional input parameters are not accepted. Returns `results`, an array of ranked candidates, and `poolSize`, the number of candidates evaluated before ranking. Optional `note` describes candidate production and `agent_hints` contains provider guidance. Candidate entries may include `paperId`, `primaryId`, source identifiers in `ids`, `title`, `authors`, `abstract`, `categories`, `createdDate`, and `updateDate`; metadata is not guaranteed to be present and some fields allow null. This is seed-based retrieval, not a guarantee of exhaustive coverage or a full-text retrieval operation. The configured synchronous SimpleMapping returns results directly; it has no separate task submission, polling, or result-fetch workflow. The provider describes `similar` as performing multi-round expansion internally, which does not imply an asynchronous client lifecycle.

ae_brightdata_get_scraper_details

View the documentation for one Bright Data scraper method before running it. Provide both required inputs: dataset_id from search_scrapers and method copied exactly from the scraper's collection_methods list, including any apparent misspellings. Returns the scraper description, input fields, example input and main output fields. Build the input for run_scraper using sample_input or input_fields. Actual collected records may contain more fields than output_fields lists.

ae_dataforseo_on_page_content_parsing

Extract structured text, headings, and links from any webpage URL using DataForSEO.

ae_bid_get_company_partners

Query the client or supplier partner list and cooperation statistics (frequency, total amount, and latest date) for a specified company based on bidding records.

ae_alphaxiv_read_files_from_github_repository

Reads files or directories from a GitHub repository, including a paper's project website source. For owner.github.io sites, try owner/owner.github.io; for owner.github.io/project/, try owner/project first. These are candidates, not guaranteed source repositories. For project videos, read the webpage first. Inspect website source only when the page omits a playable URL for a needed demo; do not recheck assets the page already exposes. Look in HTML, components, and asset manifests. Special behaviors: 1. When path is '/', returns repository structure and the contents of top-level files. Large repositories may be limited to top-level entries. 2. When path points to a directory, returns a list of files and directories, with contents of readable files fetched in parallel. Omitted files include a reason and raw URL when available. 3. When path points to a file, returns its contents. Binary files are rejected with a raw URL; PDFs should be queried with answer_pdf_queries.

ae_dumplingai_crawl_website

Crawl multiple pages of a website to capture content and metadata using DumplingAI.

ae_douyin_get_all_sec_user_id

Extract Douyin sec_user_id values from up to 10 user profile links at a time. Input: Set the required request_body parameter to an array of strings. Entries may be full profile URLs, short profile links, or copied share text containing a profile link. Results: A successfully resolved link returns its sec_user_id. If extraction fails, the original input is returned instead so you can check it manually. An unchanged link or share text does not indicate successful extraction. This tool extracts user identifiers, not profile details.

ae_dataforseo_backlinks_domain_pages_summary

Get a summary of backlinks and related metrics for each page of a domain or specific URL using DataForSEO.

ae_api_sports_football_get_fixture_lineups

Check the starting lineups and formations for a specific football fixture via API-Sports.

ae_zhihu_ai_search

Search for a topic using AI-assisted search. Parameters: - `message_content` (required): The text or question to search for. - `chat_mode` (optional, default `FAST`): The conversation mode. - `session_id` (optional, default empty): Supply a session ID to continue a multi-turn conversation. Returns a structured JSON response containing search-result data in event-list form (`events`). The response may also include status and request metadata.

ae_toutiao_search_web_content

Search Toutiao web articles by providing the required keyword string, such as a search term or query. Useful for finding related articles, researching topics, monitoring media, and collecting reference material. The response includes search data (data), a status code (code), a message (message), and a record time (recordTime). The structure of data is not specified, so particular article fields or full text are not guaranteed.

ae_xiaohongshu_resolve_share_url

Expand a Xiaohongshu (RedNote) short share link to its public redirect URL. Provide shareUrl as a URL starting with http://xhslink.com/ or https://xhslink.com/. The returned destination can be used for subsequent content lookup or processing; this tool resolves the link rather than retrieving the content itself.

ae_alphaxiv_answer_pdf_queries

Read relevant source passages from one PDF to investigate questions about a paper. Returns raw, filtered page content as XML (<paper id="..."><page num="N">...</page>...</paper>), rather than a synthesized answer. An ordinary webpage URL returns its readable text, like read_webpage; only PDFs use page ranking. Supports arXiv, alphaXiv, and Semantic Scholar abstract pages. In request_body, set paper to a paper ID, URL, or full title, and provide queries as an array with one question per string. You can supply a title without looking up its URL first, but titles must have a unique exact match; ambiguous or missing titles do not cause a substitute paper to be read. If needed, use lookup_paper to inspect candidates and then pass the chosen ID. Batch all questions about the same paper into one call. PDF calls return the most relevant pages up to a fixed size, so additional questions about that paper are nearly free. To read multiple papers, issue separate calls in parallel; they resolve as one search round. Every claim drawn from returned PDF <page> text must carry a citation tag built from that text.

ae_dataforseo_backlinks_domain_intersection

Find domains that link to your competitors but not to your site using DataForSEO. This helps identify link-building opportunities by revealing gaps in your backlink profile.

ae_dataforseo_google_search_intent

Analyze search intent for up to 1,000 keywords using DataForSEO. Classifies queries as informational, navigational, commercial, or transactional with probability scores.