The request "agent-researcher" names no specific product, so the candidates are those that do agent-driven research. Exa offers a deep research agent and an agent run for research and extraction 212, Smithery Linkup provides an autonomous research task that investigates the web 717, and Nous Research's arXiv search covers academic research 111. Strands Agents and LangChain Deepagents are open toolkits for building research agents 510.
A question back: Do you want a hosted research service to call, or a framework to build your own research agent?
- 1
Exa Search APIexa.aiverified and signed
Provides a search API for AI agents with real-time web data and deep research, and an Exa Agent run that performs deep research in the background 212.
official:MCP server · exa-mcp-serverskill · exaskill · build-with-exaskill · exa-search
for agents: deep_researcher_start · exa_create_agent_run · company-research
- 2
- 3
- 4
- 5
Sources
- 1Nous Researchnousresearch.com · read 2026-10-29
- 2Exa | Search API for AI Agents — Real-Time Web Dataexa.ai · read 2026-10-29
- 5Strands Agents | The open source toolkit for production AI agentsstrandsagents.com · read 2026-10-29
- 7Smithery - Connect agents to services in minutessmithery.ai · read 2026-10-29
- 10LangChain: the open agent platform to own your intelligencelangchain.com · read 2026-10-29
- 11arxiv: declares "arxiv" (the record we read and signed)unfragile.ai · read 2026-10-24
- 12exa-mcp-server: declares "deep_researcher_start" (the record we read and signed)unfragile.ai · read 2026-10-24
- 17linkup-mcp-server: declares "linkup-research" (the record we read and signed)unfragile.ai · read 2026-10-24
- 20deepagents: declares "agentic" (the record we read and signed)unfragile.ai · read 2026-10-24
For agents: the 10 actions behind this answer
arxiv · arxiv verified and signed
Search arXiv papers by keyword, author, category, or ID.
deep_researcher_start · exa-mcp-server verified and signed
[Deprecated] Start an AI research agent that searches, reads, and writes a detailed report. Takes 15 seconds to 2 minutes. Best for: Complex research questions needing deep analysis and synthesis. Returns: Research ID - use deep_researcher_check to get results. Important: Call deep_researcher_check
autoresearch · autoresearch verified and signed
Autonomous iterative experimentation loop for any programming task. Guides the user through defining goals, measurable metrics, and scope constraints, then runs an autonomous loop of code changes, testing, measuring, and keeping/discarding results. Inspired by Karpathy's autoresearch. USE FOR: auton
agent · @anthropic-ai/claude-agent-sdk verified and signed
hermes-agent · hermes-agent verified and signed
Use, configure, theme, extend, and orchestrate Hermes Agent.
· Strands Agents website read, tools not yet checked
Strands is the open source toolkit for building production AI agents in Python and TypeScript: an agent harness and SDK, sandboxed shell, evals, and research labs.
Find providers for a job · Agentery verified and signed
Free-text search over indexed AI agents, MCP servers, APIs and platforms with observed prices, evidenced deployment model (hosted / self-run / both / unknown) and connection details. Same handler as the MCP tool search_providers and GET /api/v1/search. e.g. search: supplier invoice reconciliation e.
linkup-research · linkup-mcp-server verified and signed
Submits an autonomous Linkup research task: an agent that investigates the web to answer questions a single search query cannot resolve, returning a synthesized, cited answer. Use for verified answers to precise questions, focused investigations of a defined subject, or broad multi-angle reports. Th
Get A2A agent · Agent Tools Directory verified and signed
Return the indexed Agent Card for one A2A agent. e.g. show me the agent card for slug acme-research
· AntiNode- Your own research agent website read, tools not yet checked
AntiNode AI is a powerful tool that allows users to perform report analysis, web scraping, multi document analysis solo as well with their teams or friends
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.