construct-evidence-network
SkillDev toolsLets your agent build an evidence graph comparing treatments and check whether the studies can be validly combined.
Available today. Use it from your connected AI after setup.
No other account needed.
Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
Then ask your AI: use the construct-evidence-network skill
About this skill
Construct a network-meta-analysis evidence graph and evaluate connectivity/transitivity prerequisites.
What this skill tells your AI
The instructions your AI receives, as published by yogsoth-ai/de-anthropocentric-research-engine in skills/construct-evidence-network/SKILL.md and read by ahel’s review.
Purpose
Construct an evidence graph for network synthesis and evaluate connectivity, transitivity, and comparison support.
Input contract
required: [evidence_records, intervention_nodes, outcome_schema]
optional: [study_designs, effect_sizes, covariate_schema]
constraints: [edges require a shared outcome definition and traceable comparison evidence]
Procedure
- Normalize interventions, comparators, outcomes, and study-level edges.
- Build the direct and indirect comparison graph with effect-support provenance.
- Check connectivity and transitivity assumptions across connected components.
- Report usable components, unsupported links, and synthesis limitations.
Output contract
produces: [evidence_graph, connected_components, transitivity_assessment, unsupported_links, synthesis_readiness]
delta_fields: [findings, evidence_updates, uncertainties, decisions, open_questions]
Quality gates
- Every edge points to a study record and outcome definition.
- Disconnected components are not silently combined.
- Transitivity judgments list the effect modifiers examined.
Failure and counterexamples
Do not create indirect evidence from incomparable populations or outcomes, and do not treat graph connectivity as proof of exchangeability.
Provenance map
resolved: knowledge-acquisition-evidence-network-construction
Signals
- GitHub stars
- 501
- Forks
- 41
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Key
construct-evidence-network- Source
- github.com/yogsoth-ai/de-anthropocentric-research-engine