BGPT MCP
VerifiedMCP server delivering full-text scientific paper data to AI agents and Python apps.
What is BGPT MCP?
BGPT MCP functions as a remote server exposing scientific literature search through the Model Context Protocol and a standard JSON API. It processes queries against full-text studies to extract granular fields such as results, conclusions, conflicts of interest, and falsifiability prompts.
Integration works by pointing MCP clients at hosted endpoints or running the npm package locally. Results include up to 25 metadata elements per paper, enabling agents to evaluate evidence strength before summarizing findings.
The service targets researchers, literature-review agents, and developers building evidence dashboards who need deeper paper content than typical search engines supply.
What you can build with BGPT MCP
Evidence interrogation
Feed a scientific claim into the search tool to retrieve studies with explicit limitations and sample details for balanced analysis.
Literature dashboard creation
Generate interactive Plotly reports that surface methods, data availability, and blind spots across multiple papers on a topic.
Agent-assisted reviews
Connect directly to Claude or Cursor so the AI can pull structured paper data during real-time research tasks.
Install BGPT MCP
npm install -g bgpt-mcpimport requests
def search_bgpt(query, num_results=10, days_back=None, api_key=None):
payload = {"query": query, "num_results": num_results}
if days_back is not None:
payload["days_back"] = days_back
if api_key:
payload["api_key"] = api_key
response = requests.post(
"https://bgpt.pro/api/mcp-search",
json=payload,
timeout=30,
)
response.raise_for_status()
return response.json()["results"]
papers = search_bgpt("CRISPR delivery neurons", num_results=5)
print(papers[0]["title"])- 1Choose an MCP client such as Claude Desktop or Cursor.
- 2Add the remote SSE endpoint https://bgpt.pro/mcp/sse to your client config file.
- 3Restart the client to load the BGPT server connection.
- 4Run a test query like search_papers with your research topic and desired result count.
- 5Review returned papers for experimental details and metadata fields.
BGPT MCP: pros & cons
Pros
- +Provides granular experimental data beyond titles and abstracts
- +Works with multiple popular AI clients via standard MCP transports
- +Free tier available with no API key required for basic use
- +No local infrastructure or Docker setup needed
Cons
- –Advanced usage may require a paid Stripe subscription key
- –Search scope limited to the provider's curated paper database
- –Remote dependency means availability tied to bgpt.pro uptime
Frequently asked questions
The free tier works without a key for up to 50 results; paid access uses a Stripe subscription ID.
User reviews
Verified reviews from the community shape this listing's rating.
Loading reviews…
Similar agents
Other general-purpose options worth comparing.
