Other

Scientific Consensus

scientific-consensus-pp-cli

Aggregate PubMed, OpenAlex, Crossref, and Europe PMC into evidence summaries, consensus scores, and study-design pyramids no single search tool produces.

Printed by @laci141

Install

npx — recommended

npx -y @mvanhorn/printing-press-library install scientific-consensus

Installs the CLI binary and the agent skill (Claude Code, Codex, Cursor, Gemini CLI, Copilot, and more). Add --cli-only or --skill-only for one half. Requires Node.

Hermes

hermes skills install mvanhorn/printing-press-library/cli-skills/pp-scientific-consensus --force

From the Hermes CLI. Inside a chat session use /skills install with the same path.

OpenClaw

Install the pp-scientific-consensus skill from https://github.com/mvanhorn/printing-press-library/tree/main/cli-skills/pp-scientific-consensus. The skill defines how its required CLI can be installed.

Paste this to your OpenClaw agent — the skill describes how to install its CLI.

Claude Desktop

This CLI ships an MCPB bundle. Download the latest release .mcpb and double-click it — Claude Desktop walks you through the install.

View source on GitHub →

Aggregate PubMed, OpenAlex, Crossref, and Europe PMC into evidence summaries, consensus scores, and study-design pyramids no single search tool produces.

Scientific Consensus turns large collections of papers into actionable evidence. It scores consensus across sources (consensus), classifies studies by design and renders evidence pyramids (evidence), detects gaps and controversies, and persists everything to a local SQLite store you can query offline with --json. Fully keyless; optional AI keys upgrade summarization.

Learn more at Scientific Consensus.

Created by @laci141 (laci141).

Authentication

No API key required for any command. Optional env vars raise limits or enable AI summarization: NCBI_API_KEY (PubMed, higher rate limit), SEMANTIC_SCHOLAR_API_KEY (Semantic Scholar enrichment), and OPENAI_API_KEY / ANTHROPIC_API_KEY / GEMINI_API_KEY (enhanced summarization). Everything works without them.

Quick Start

# Verify per-source reachability before running analyses.
scientific-consensus doctor --dry-run

# Cross-source search to confirm data flows.
scientific-consensus search "vitamin d covid" --limit 10

# The headline command: an evidence-backed consensus verdict.
scientific-consensus consensus "vitamin D reduces respiratory infections"

# See the evidence pyramid for the same topic.
scientific-consensus evidence "vitamin d covid"

Unique Features

These capabilities aren't available in any other tool for this API.

Evidence intelligence

consensus — Answer 'what does the evidence say about X' with a Consensus Score, Confidence Score, and Evidence Strength across all sources.

Reach for this when an agent needs an evidence-backed yes/no/mixed verdict instead of a raw paper list.
```
scientific-consensus consensus "vitamin D reduces respiratory infections" --agent
```
evidence — Classify retrieved studies by design (meta-analysis to case report) and render the evidence pyramid for a topic.

Reach for this to judge whether a claim rests on RCTs/meta-analyses or just case series.
```
scientific-consensus evidence "intermittent fasting weight loss" --agent
```
compare — Run two consensus analyses side-by-side to compare competing claims or interventions.

Reach for this when an agent must weigh two interventions or contradictory claims.
```
scientific-consensus compare "statins reduce mortality" "statins increase diabetes risk" --agent
```
reproducibility — Estimate reproducibility by detecting replication studies, sample sizes, and pre-registration cues.

Reach for this to gauge how well-replicated a finding is.
```
scientific-consensus reproducibility "power posing" --agent
```
quality — Estimate overall study quality from design, venue prestige, sample-size cues, and citation mass.

Reach for this for a quick quality signal before deep reading.
```
scientific-consensus quality "omega-3 depression" --agent
```

Discovery

gaps — Identify understudied populations, missing long-term/replication/RCT studies, and future directions for a topic.

Reach for this to find what research is missing, not just what exists.
```
scientific-consensus gaps "pediatric long covid" --agent
```
controversies — Surface conflicting studies, contradictory conclusions, and rapidly changing evidence for a topic.

Reach for this when the question is 'is this settled or disputed?'
```
scientific-consensus controversies "saturated fat heart disease" --agent
```
funding — Analyze funding patterns and funder concentration for a research topic.

Reach for this to see who funds research on a topic (potential conflicts).
```
scientific-consensus funding "e-cigarette safety" --agent
```

Trends

emerging — Detect the fastest-growing research areas and exploding publication trends.

Reach for this to spot hot research areas before they peak.
```
scientific-consensus emerging --field neuroscience --agent
```
drift — Compare a field's topic distribution between two year windows to spot emerging and fading subtopics.

Reach for this to see how a field's focus shifted over time.
```
scientific-consensus drift "machine learning genomics" --from 2015 --to 2025 --agent
```
watch — Monitor a topic and report major new publications since the last run.

Reach for this to keep an agent or researcher current on a fast-moving topic.
```
scientific-consensus watch "GLP-1 cardiovascular outcomes" --agent
```

Recipes

Evidence-backed verdict for an agent

scientific-consensus consensus "creatine improves cognition" --agent --select verdict,consensus_score,confidence,study_count

Returns a compact JSON verdict an agent can act on without parsing papers.

Evidence pyramid as a table

scientific-consensus evidence "mediterranean diet cardiovascular" --csv

Study-design distribution from meta-analyses down to case reports.

Compare two claims

scientific-consensus compare "intermittent fasting weight loss" "calorie counting weight loss" --agent

Side-by-side consensus and evidence strength for competing approaches.

Curate a deduped reading list

scientific-consensus curate "crispr off-target effects" --format bibtex --limit 25

Ranked, cross-source, DOI-deduplicated reading list exported as BibTeX.

Track a fast-moving topic

scientific-consensus watch "GLP-1 cardiovascular outcomes" --agent

Reports new publications since the last run from the local baseline.

Usage

Run scientific-consensus-pp-cli --help for the full command reference and flag list.

Commands

authors

Search and retrieve authors

scientific-consensus-pp-cli authors get - Get a single author by OpenAlex ID
scientific-consensus-pp-cli authors search - Search authors

funders

Research funders

scientific-consensus-pp-cli funders - Search funders

institutions

Search and retrieve institutions

scientific-consensus-pp-cli institutions get - Get a single institution by OpenAlex ID
scientific-consensus-pp-cli institutions search - Search institutions

sources

Journal (source) metadata

scientific-consensus-pp-cli sources get - Get a journal (source) by ISSN or OpenAlex ID
scientific-consensus-pp-cli sources search - Search sources (journals)

topics

Research topics

scientific-consensus-pp-cli topics - Search topics

works

Search and retrieve scholarly works

scientific-consensus-pp-cli works get - Get a single work by OpenAlex ID, DOI, or PMID
scientific-consensus-pp-cli works search - Search scholarly works

Output Formats

# Human-readable table (default in terminal, JSON when piped)
scientific-consensus-pp-cli authors get mock-value

# JSON for scripting and agents
scientific-consensus-pp-cli authors get mock-value --json

# Filter to specific fields
scientific-consensus-pp-cli authors get mock-value --json --select id,name,status

# Dry run — show the request without sending
scientific-consensus-pp-cli authors get mock-value --dry-run

# Agent mode — JSON + compact + no prompts in one flag
scientific-consensus-pp-cli authors get mock-value --agent

Agent Usage

This CLI is designed for AI agent consumption:

Non-interactive - never prompts, every input is a flag
Pipeable - --json output to stdout, errors to stderr
Filterable - --select id,name returns only fields you need
Previewable - --dry-run shows the request without sending
Read-only by default - this CLI does not create, update, delete, publish, send, or mutate remote resources
Offline-friendly - sync/search commands can use the local SQLite store when available
Agent-safe by default - no colors or formatting unless --human-friendly is set

Exit codes: 0 success, 2 usage error, 3 not found, 5 API error, 7 rate limited, 10 config error.

Health Check

scientific-consensus-pp-cli doctor

Verifies configuration and connectivity to the API.

Configuration

Config file: ~/.config/scientific-consensus-via-pp-cli/config.toml

Static request headers can be configured under headers; per-command header overrides take precedence.

Troubleshooting

Not found errors (exit code 3)

Check the resource ID is correct
Run the list command to see available items

API-specific

Semantic Scholar results missing or sparse — Semantic Scholar rate-limits keyless requests (HTTP 429). Set SEMANTIC_SCHOLAR_API_KEY or rely on OpenAlex/PubMed/Europe PMC; commands degrade gracefully.
PubMed throttling on large syncs — Set NCBI_API_KEY to raise the rate limit from 3 to 10 requests/second.
consensus reports method=heuristic — Stance classification is lexical without an AI key. Set OPENAI_API_KEY, ANTHROPIC_API_KEY, or GEMINI_API_KEY for AI-assisted stance detection.
Empty results offline — Run 'scientific-consensus sync ""' first to populate the local store, or use --data-source live.

Sources & Inspiration

This CLI was built by studying these projects and resources:

semanticscholar — Python (400 stars)
pyalex — Python (350 stars)
habanero — Python (250 stars)
metapub — Python (200 stars)
europepmc — R (150 stars)

Generated by CLI Printing Press