</>MCP Agents Market
Agent

hyperresearch

by jordan-gibbs1.8kPythonUpdated 2026-08-04

Agent-driven research knowledge base. Agents collect, search, and synthesize web research into a persistent, searchable wiki.

Claude CodeClaude DesktopCursor

Hyperresearch is an AI sub-agent that converts Claude Code into a comprehensive deep research platform with a tier-adaptive 16-step pipeline. It autonomously collects, verifies, and synthesizes web and scholarly research into adversarially-audited reports while maintaining every source in a persistent, searchable markdown-and-SQLite vault. The system processes 100-250+ sources per run, performs citation verification, detects syndicated content, and supports research scales from 30-minute queries to multi-chapter dissertations spanning 300-450 sources. All research accumulates in a reusable knowledge base accessible through CLI, MCP server, or web interface.

Key Features

16-step research pipeline with automatic tier routing (light, full, dissertation) and configurable scale profiles targeting 55-450 sources
Adversarial verification system with four parallel critics, citation integrity checks, retraction detection, and surgical-only editing constraints
Persistent markdown-and-SQLite vault that compounds across sessions with full-text search, semantic search, PageRank ranking, and provenance tracking
Unified scholarly discovery client querying OpenAlex, Crossref, CORE, DOAB, ClinicalTrials.gov, SEC EDGAR, and FRED with DOI-based deduplication
Automatic open-access full-text recovery through Unpaywall, Europe PMC, and CORE when sources are paywalled, with version-aware substitution disclosure
Resumable runs with per-step manifests, budget caps, and concurrent workspace isolation
Authenticated browser crawling with escalation queue for login-walled content and Claude-in-Chrome integration
Multi-interface access via Claude Code skill, MCP server (13 tools), CLI, and local web UI with no JavaScript dependencies

Use Cases

  • 01Conducting comprehensive literature reviews across 100+ sources with automated citation verification and adversarial fact-checking
  • 02Building a persistent research knowledge base that grows smarter across sessions by reusing previously fetched sources
  • 03Writing long-form research reports or dissertation chapters from 300-450 sources with automatic open-access paper recovery
  • 04Querying scholarly databases, clinical trials, SEC filings, and economic data through a unified interface with intelligent deduplication
  • 05Investigating contradictory claims across sources with independence auditing to weight syndicated content appropriately
  • 06Resuming interrupted deep research sessions exactly where they stopped without re-fetching or re-analyzing sources

Related Agents

View more

hermes-agent

The agent that grows with you

Agent250.4k

agency-agents

A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.

Agent155.5k

openinterpreter

A coding agent for open models like Kimi K3

Agent68k

cline

Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Agent66.3k

hyperresearch — FAQ

What is hyperresearch?+

Hyperresearch is an AI sub-agent that transforms Claude Code into a deep research system. It runs a 16-step pipeline that collects, verifies, and synthesizes research from web and scholarly sources into adversarially-audited reports while maintaining all sources in a persistent, searchable vault.

How do I install the hyperresearch AI sub-agent?+

Install via pip with `pip install hyperresearch && hyperresearch install` in your project directory, then invoke `/hyperresearch <your research query>` in Claude Code. Requires Python 3.11-3.13; use `hyperresearch install --global` to make it available across all Claude Code sessions.

Which AI clients work with hyperresearch?+

Hyperresearch works primarily with Claude Code as a native skill. It also provides an MCP server (`pip install hyperresearch[mcp]` then `hyperresearch mcp`) that works with Claude Desktop, Cursor, and any MCP-compatible client for vault access and search.

What API keys or prerequisites does hyperresearch need?+

The core system works with Claude Code's built-in Anthropic access with no additional keys. Optional enhanced features require: CORE_API_KEY for CORE scholarly search, FRED_API_KEY for economic data, contact email for Unpaywall open-access recovery, and API keys for Exa or Tavily if using those web providers instead of the default.

Is hyperresearch free to use?+

The software is MIT-licensed and free. Usage incurs Anthropic API costs through Claude Code based on research tier (light runs ~30-40 min, full runs ~1.5-2.5 hours, premier profile ~3-5 hours, dissertation runs ~4-8 hours). Budget caps can be set with `hyperresearch run init --budget <amount>`.

How does the persistent vault work across research sessions?+

Every fetched source lands in a markdown-and-SQLite vault at `research/notes/`. Future sessions search the vault before fetching anything new, so research compounds over time. The vault is plain markdown with YAML frontmatter, fully rebuildable, and accessible via CLI, MCP, web UI, or any text editor.

How do I install hyperresearch?+

Open the source repository on GitHub and follow its README. hyperresearch is a agent — MCP Agents Market links you directly to the official repo.

Is hyperresearch free?+

hyperresearch is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches

Research AgentsKnowledge & Memory AgentsDeveloper Tools AgentsBest MCP ServersBest SkillsBest PluginsTop rated listingsClaude Code toolsBrowse all categories