Live web search, news from 45+ countries, full article text and research papers for AI agents.
Scored across 9 tools
The vertical directories (ai_tools, games, software, research_papers) are clearly distinct, but the three news entry points (news by topic, country_news by country, news_search by keyword) overlap enough that an agent must read the descriptions carefully to pick correctly. web_search also mixes in fresh news and QA/Wikipedia cards, adding a second overlap zone. The descriptions do explicitly cross-reference each other (country_news points to news_search), which keeps this from being worse.
All names use snake_case, so casing is uniform and predictable. The pattern itself drifts, however: bare nouns (news, games, software), noun_noun compounds (country_news, news_search, web_search, research_papers, ai_tools), and one verb_noun (read_page). Readable and mostly consistent, with the read_page verb form being the only real outlier.
Nine tools is well-scoped for a content-aggregation server: one tool per content vertical plus two cross-cutting utilities (web_search, read_page). Nothing feels redundant in count and no vertical is left without an entry point.
The surface covers discovery across news, web, papers, AI, games and software, with full_text/read_page to go deeper, so the main agent workflows are supported. Minor gaps remain: there is no way to fetch a single item's details by id (a specific game, tool or paper) short of a targeted search, and read_page handles only one URL per call.
AI models, apps and open-source projects, updated daily: every new model on OpenRouter (with prices per million tokens), trending Hugging Face models and apps, fast-growing GitHub AI / agent / MCP projects, and the big assistants with their newest models. query: optional words (name, use or vendor, e.g. "video", "claude", "mcp server"). category: all, llm, image, video, audio, code, writing, design, 3d, data, agents, education. sort: latest (newest releases first), trending (most used or starred) or name. limit: 1-50.
Inputs: sort, limit, query, category. Returns: result
Latest local news from one country. country: two-letter code, e.g. MY, SG, ID, TH, PH, KR, US, GB, JP, IN, AU. category: all, breaking, business, sports, entertainment, politics, health or science. limit: 1-50. full_text: true to also get the full text of the first 5 articles (each read counts as one extra API call). To find news about a subject, use news_search with country.
Inputs: country (required), limit, category, full_text. Returns: result
Games, updated daily from Steam: top sellers, new releases, coming soon, deals and the most played right now, with description, genre, platforms, price, release date, rating (0-10 from player reviews) and players_now. query: optional words (title, genre or developer). genre: all, action, adventure, rpg, strategy, sports, racing, puzzle, simulation. sort: trending (most played now), latest (newest releases), rating or name. limit: 1-50.
Inputs: sort, genre, limit, query. Returns: result
Latest news headlines with links and summaries, refreshed every few minutes. topic: global, tech or financial. category: a section inside the topic (e.g. 'ai', 'crypto', 'stocks', 'politics'), or 'all'. limit: 1-50 (default 10). full_text: true to also get the full text of the first 5 articles (each article read counts as one extra API call). To find news about a subject, use news_search.
Inputs: limit, topic, category, full_text. Returns: result
Find news articles about anything by key words: every article NestDaddy collected from 45+ countries in the last 120 days, best match and newest first. The best tool for questions about recent events, people, companies and places. Use the words a headline would use, in the local language for local news (e.g. "Najib perintah pelepasan"). country: optional two-letter code (MY, SG, US ...). days: 1-120 (default 30). limit: 1-50 (default 10). full_text: true to also get the full text of the first 5 articles (each article read counts as one extra API call).
Inputs: query (required), days, limit, country, +1 more. Returns: result
Read the full text of one web page or news article (title, author, date and the main text, without menus or ads). Use it after news or web_search when you need the details behind a headline. url: the page link from a search or news result. max_chars: 500-30000 (default 8000).
Inputs: url (required), max_chars. Returns: result
Scientific papers from arXiv, PubMed, CORE, OpenAlex, CrossRef, Europe PMC and bioRxiv. Returns title, authors, year, abstract and link. limit: 1-50 (default 10).
Inputs: query (required), limit. Returns: result
Software and developer tools with their current version, updated daily (Homebrew installs of the last 30 days): name, description, category, platform, latest_version, install command (in tags) and link. query: optional words. category: all, development, productivity, design, security, utilities, multimedia, browser, communication, ai, cloud. sort: trending (most installed), latest (recently updated) or name. limit: 1-50.
Inputs: sort, limit, query, category. Returns: result
Search the web with NestDaddy's own index: title, url, snippet and domain of each page. For time-sensitive
queries (a year, "latest", "today", "harga ...") fresh news articles are mixed in on top (type "news").
Official sources are mixed in too: Stack Overflow answers for programming errors (type "qa"), Wikipedia for
people, places and "what is" questions (type "wiki"), and an answer card with live official numbers for
Malaysian fuel prices (data.gov.my), gold and exchange rates (Bank Negara) and weather (MET Malaysia).
Use short key words, not a whole sentence. Read the best pages with read_page before explaining details.
query: what to search for. limit: 1-50 results (default 10).
Inputs: query (required), limit. Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Agent-native search engine with live web research optimized for AI agents.
Live web search, scraping and crawling API for AI. Get structured, LLM-ready data from any site.
News search for AI agents: recent news on any topic from outlets worldwide, by country and language.
The best web search for your AI Agent
Add one secure layer between your agents and this server.