Search scientific papers by topic or DOI, a paper’s identifier, and organize evidence extracted from full text. Connect claims with experiments, reported statistics, limitations, and source locations for literature reviews and evidence tables.
Searches BGPT scientific papers by topic or DOI and retrieves claim-level evidence extracted from full text, including experiments, reported statistics, scope, limitations, and provenance. Use for literature reviews, evidence synthesis, and finding experimental details beyond abstracts.
Run score Not tested
See how to installThis skill searches for scientific papers by topic or DOI, a paper’s identifier, and organizes claims extracted from full text with their supporting experiments and reported statistics. It also helps review the scope, limitations, and source locations behind those claims, and build comparison tables that let readers trace the evidence. Searches can be filtered by paper type and publication date.
Use it when preparing a literature review or looking for experimental details that are not included in abstracts. It can also help compare multiple studies and organize where their findings agree or differ.
A remote BGPT MCP connection and internet access are required, and installing the skill does not configure that connection. A local bridge additionally needs Node.js and npm. Paid use requires a BGPT API key. Returned records are extracted evidence, not necessarily the full text; some may contain no evidence. Verify important numbers against the original paper. Do not treat the search as exhaustive or as an assessment of research bias. Check pricing and any free allowance before retrieving many results.
This overview was written by AI (OpenAI Codex) from the public SKILL.md. Check the original below for the exact behavior.
AI checked the SKILL.md for operations that need care and for setup requirements. The score comes only from actually running the skill (hands-on test).
Not counted in the score. Rated by AI (OpenAI Codex) as of Oct 24, 2026. Treat it as a guide and read the original before installing.
Review the instructions the agent will receive before you install.
Use BGPT to find full-text-derived evidence for a research question or a known DOI. It returns extracted claims and their supporting experiments, rather than only bibliographic metadata. Use these records to build a traceable evidence table; they are not a complete systematic-review search or a validated risk-of-bias assessment.
Configure BGPT in the host before use; installing this skill does not register an MCP server. The provider supports both transports:
| Transport | Endpoint |
|---|---|
| Streamable HTTP | https://bgpt.pro/mcp/stream |
| SSE | https://bgpt.pro/mcp/sse |
Use the host's native remote-MCP configuration when supported. For a host that
uses mcpServers JSON and requires a local stdio bridge, this configuration uses
mcp-remote (adapt it to the host's actual settings format):
{
"mcpServers": {
"bgpt": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://bgpt.pro/mcp/stream", "--transport", "http-only"]
}
}
}
The published bgpt-mcp 1.1.0 npm package is a thin SSE wrapper around
mcp-remote, not a local paper database. Its bundled README predates the current
tool contract and pricing. Prefer the native connection or configurable bridge
above. The linked BGPT GitHub repository was unavailable at review time; use
the current provider documentation and live tool schema.
Free access needs no key. For paid usage, configure Authorization: Bearer <key>
in the host's remote-server headers, using its credential storage. For the bridge,
mcp-remote supports --header-file
with a local credential file. Do not put api_key in an MCP tool call: neither
current tool declares that argument. Authenticate through the connection.
Discover the connected server's tools before calling them. The live BGPT 2.14.7 schema reviewed on 2026-09-30 exposes:
| Tool / argument | Contract |
|---|---|
search_papers.query | Required string. Use short English search terms. |
num_results | Integer, 1–100; default 16. Set it explicitly to control result use. |
days_back | Optional integer; papers published within the last N days. Omit for all dates. |
min_citations | Optional integer; minimum number of references cited by the paper, not citations received. |
study_type | Optional string: primary study, systematic review, meta-analysis, narrative review, protocol, dataset, commentary, or other. |
output_format | evidence (default), full, or legacy. |
lookup_paper.doi | Required DOI string; also accepts output_format. A found paper counts as one result. |
Use the filter arguments instead of putting years or filters into the query. No Boolean-query grammar, offset, page, or cursor is documented by this schema. A result limit is not pagination: do not invent continuation arguments or claim the search enumerates every matching paper. For broader coverage, run explicit question variants and deduplicate by DOI while retaining the query history.
Search example verified with one public result (invoke through the connected tool interface):
{"query": "CRISPR human cells", "num_results": 1, "output_format": "full"}
Illustrative filtered search (filters were schema-checked, not executed):
{"query": "CRISPR delivery neurons", "num_results": 5, "days_back": 90, "study_type": "primary study", "output_format": "evidence"}
For a paper already identified, call lookup_paper with its bare DOI rather
than hoping a title search returns that exact paper:
{"doi": "10.1016/j.ymeth.2015.10.014", "output_format": "evidence"}
Do not interpret a connection, quota, or tool error as an empty evidence set. The provider publishes a result allowance, but no numeric request-rate limit on the reviewed page; avoid assuming unlimited request throughput.
Search returns an envelope with a results list and count; record the returned
query too, since the service may rewrite search terms. DOI lookup returns found
and a single result; handle not-found before reading it. Inspect the MCP
result for tool errors before interpreting the envelope.
evidence gives compact evidence records; empty optional fields may be absent.full adds legacy paper metadata to the evidence record.legacy requests the older paper metadata representation. Do not assume every
paper has all previously advertised fields or a quality score.The provider says evidence/full formats omit papers without extracted evidence.
However, the live smoke test returned a record with schema_version: "legacy",
extraction_status: "legacy", and an empty evidence_units list in both formats.
A returned record therefore does not guarantee claim-level provenance. An empty
search or unfound DOI also does not establish that the paper does not exist.
Retain doi, title, publication_date, and publication_name when available,
then inspect central_claim and evidence.evidence_units. Each unit links its
claim to an experiment, reported statistics, demonstrated scope, and provenance.
Keep the schema version: older records can have a different shape. Legacy full
metadata can contain nulls, numeric strings, and serialized text instead of
native lists; preserve the raw values and never execute them as code.
The provider documents V4 limits of five evidence units per record, two provenance
passages per unit, and 320 characters per passage. These are bounded extractions,
not the complete article. Check extraction_status, normalization_warnings,
and evidence.record_provenance, including truncation information, before synthesis.
For each extracted result, retain the DOI and source section, table, or figure. Verify sample sizes, units, experimental arms, reported statistics, and uncertainty against the paper before using them in a synthesis. Distinguish author-reported limitations from extraction-generated interpretations. Preserve contradictory or mixed evidence and demonstrated scope. If legacy quality scores are present, use them only as screening aids, not as a study-specific risk-of-bias assessment.
At the review date the provider advertises 50 free results per network, then $0.02 per returned result. A successful DOI lookup counts as one result; unfound lookups cost none. Recheck the allowance and pricing before a large batch. Do not repeatedly retry a quota-exhausted request.
Sources: provider setup, record formats, and billing, live MCP endpoint, published BGPT wrapper metadata, and bridge documentation. The tool schemas, one-result full search, matching evidence-format DOI lookup, and SSE handshake were checked with unauthenticated requests. The public search/lookup returned a legacy-schema record, not a V4 fixture. Paid authentication, filtered search behavior, and host-specific configuration were not tested.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Estimate intracellular reaction rates from steady-state carbon-13 tracing measurements. Validate carbon atom mappings and measurement inputs, simulate labeling, fit constrained fluxes, and diagnose which pathway rates the data can meaningfully constrain.
Plan protein characterization experiments with Adaptyv's cloud laboratory, estimate costs, submit sequences, track progress, and retrieve measured results. Supports binding, thermal stability, expression, fluorescence, and enzyme activity workflows through the API and Python SDK.
Analyze time series with aeon, a Python toolkit for data that changes over time. Supports classification, regression, clustering, forecasting, anomaly detection, and pattern search, with guidance on preprocessing, algorithm selection, and evaluation.
Investigate how DNA variants may affect gene activity using precomputed AlphaGenome Atlas results and model predictions. Rank candidate variants, examine tissue-specific effects, and inspect the features behind scores to develop research hypotheses about biological mechanisms.
Plan and document studies that assess whether an analytical procedure suits its intended use. Select the governing framework, evaluate calibration, accuracy, precision, and detection limits, and compare methods using acceptance criteria defined before data collection.
Organize single-cell measurements with cell and gene metadata, read and write h5ad or Zarr files, and prepare subsets or combined datasets. Includes guidance on handling large annotated matrices without loading all measurement values into memory.