# Research Scope

> Research Scope is an open-source MCP server for academic research. MCP server that plans research, scores sources and checks report citations. It is available on CLI, API and self-hosted.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, API, self-hosted
- **Website:** https://axelverne.com
- **Repository:** https://github.com/axelvernebooks/research-scope-mcp
- **GitHub stars:** 0
- **License:** MIT
- **MCP Registry name:** io.github.AxelVerneBooks/research-scope-mcp
- **Categories:** [academic research](https://ai-powered.com/categories/academic-research), [agent tooling](https://ai-powered.com/categories/agent-tooling)
- **Listing updated:** 2026-11-18

## About

Research Scope gives an MCP client a research-to-report pipeline. A planning step turns a scope into a source-target plan with kill criteria, so the workflow stops and reports insufficient sources when the evidence is too thin to continue. Each source is crawled and given a 0-100 score, with the evidence for primary-source status, freshness, paywall and conflict-of-interest flags returned alongside the number.

Reports are assembled as a citation registry with numbered references and a scoring appendix. A check step then verifies that every substantive claim cites at least one scored source and returns orphan claims as errors rather than a pass. The server runs over stdio or Streamable HTTP and ships with a Dockerfile and client config snippets.

It suits people who use AI agents for literature or source-based research and need reports where a reader can trace each sentence to a source. The MIT-licensed core is free, and the README mentions a separate paid kit with a field guide and example packs.

## Features

- **Source-target planning:** Turns a research scope into a plan with kill criteria that halt the run when sources are insufficient.
- **Explainable source scoring:** Scores each source from 0 to 100 across availability, primary status, freshness, access and independence, with the evidence returned.
- **Citation registry reports:** Builds report shells with a claims section, numbered citation registry and scoring appendix.
- **Claim-level citation checks:** Flags any substantive claim that lacks a scored source and returns orphan claims as errors.
- **Stdio and HTTP transports:** Runs locally over stdio or remotely over Streamable HTTP with unauthenticated requests answered with discovery headers.

## Good for

- Vet a set of web sources for primary status and freshness before writing a report
- Produce a research summary where every claim links to a numbered source
- Stop an agent from drafting conclusions when there are too few independent sources
- Audit an agent-written report for uncited claims before sharing it

## Tools

- `plan_research`: Turns a research scope into a source-target plan with kill criteria.
- `fetch_and_score`: Crawls one source and returns a 0-100 quality score with flags and evidence.
- `build_report`: Assembles a citation-registry report shell for scored sources.
- `check_report`: Verifies that every substantive claim cites at least one scored source.

## FAQ

### What does the scoring rubric measure?

The README lists five dimensions: availability and substance (20), primary-source status (32), freshness (24), access (12) and independence (12). Scores of 80 or more map to grade A, 65 or more to B, 45 or more to C, and anything lower to D.

### How do I run the server?

The README shows npx research-scope-mcp for stdio clients and a --http flag with a port for remote deployments. A Docker build and run command is also included.

### Is it free?

The README says the server is MIT-licensed and describes it as the free, tested core. It also mentions a separate paid kit with a field guide, worked example packs and setup support.

### Is Research Scope free?

Yes. Research Scope is open source, so you can use and self-host it for free.

### What are the best alternatives to Research Scope?

The closest alternatives to Research Scope are WritersLogic research agents, Research Mcp, Arxiv MCP Server and WritersLogic verifiable AI agents.

## Where this comes from

- Description: written from the project's README, the project's website and its MCP Registry entry.

## Alternatives

- [WritersLogic research agents](https://ai-powered.com/mcp/writerslogic-research-agents): WritersLogic research agents is an MCP server for academic research.
- [Research Mcp](https://ai-powered.com/mcp/research-mcp-3): Research Mcp is a freemium MCP server for academic research.
- [Arxiv MCP Server](https://ai-powered.com/mcp/arxiv-mcp-server): Arxiv MCP Server is an open-source MCP server for health and science.
- [WritersLogic verifiable AI agents](https://ai-powered.com/mcp/writerslogic-verifiable-ai-agents): WritersLogic verifiable AI agents is a paid MCP server for agent tooling.

---
Source: https://ai-powered.com/mcp/research-scope
