literature-consensus-map

name	literature_consensus_map
description	Summarize the current literature consensus, disagreements, and evidence strength for a biological question.
category	bio/literature
version	1
requires_tools	["search_knowledge_base","evidence_retrieval","evidence_review","python_repl"]
requires_network	true
user_invocable	true
tags	["literature","consensus","controversy","evidence-synthesis"]
aliases	["consensus_builder","literature_synthesis"]
species	any
modality	literature
stage	interpretation
stability	stable
safety_level	medium

Literature Consensus Map

Purpose

Build a concise map of what the literature broadly agrees on, where disagreement remains, and how strong the evidence appears to be.

When to use

Use this skill when the user asks for the current state of the field on a question, such as whether a pathway drives a phenotype or whether a target is well supported.

Required inputs

• question: the biological claim, mechanism, or target of interest
• context (optional): species, tissue, disease, or experimental system

Steps

1. Restate the biological question and the key scope assumptions such as species, tissue, disease, and perturbation context.
2. Search knowledge/ with search_knowledge_base for local notes, prior evidence syntheses, or project-specific framing.
3. Use evidence_retrieval to gather a bounded set of relevant PubMed-backed evidence cards for the question.
4. Use evidence_review to separate supported conclusions, unresolved conflicts, and explicit unsupported claims when the evidence is thin.
5. Use python_repl if helpful to organize the included studies into a small comparison table.
6. Keep citations attached to claims and do not overstate consensus when the literature is sparse or context-dependent.

Output format

• Biological context or assumptions: species, system, disease, and scope used to judge consensus.
• Evidence or source basis: which search_knowledge_base, evidence_retrieval, and evidence_review artifacts supported the summary.
• Consensus map: strong recurring claims, context-dependent findings, and disagreement or uncertainty.
• Caveats or ambiguity: weak evidence base, unresolved conflicts, or questions that remain unsupported.
• Recommended next step: what evidence gap to resolve, experiment to prioritize, or narrower follow-up question to ask.

Failure modes

• Too broad a question: ask the user to narrow the scope.
• Sparse literature: say the consensus is weak or preliminary.
• Mixed systems: separate conclusions by species or model rather than collapsing them.

Examples

• "What is the literature consensus on TOX in T-cell exhaustion?"
• "Summarize the evidence for interferon signaling in melanoma resistance."