clearai-dsh
View on GitHubClearAI is a native DSH plugin that brings the Epistemic Loop to DeepSeek Harness.
ClearAI is a DSH (DeepSeek Harness) agent preset that runs a seven-stage epistemic loop — frame, hypothesize, plan, observe, verify, evaluate, record — to grow a verifiable domain ontology. Each knowledge edge needs evidence and independent evaluation, with contradictions surfaced for human decision.
Use Cases
Build a growing domain ontology from research sessionsTrack hypotheses with evidence chains and support levelsSurface contradictory conclusions between research branchesAI for Science experiment reasoning (sensor drift, WENO convergence)Keep numerical evidence separate from mathematical proofRun a verifier-separated agent loop where the doer cannot self-judgeRetain refuted hypotheses instead of discarding themUse as a ClearAI preset mode inside dsh web sessionsManage skills and external memory catalog in one placeInspect knowledge graph nodes, edges, and revision history
Built With
- Language
- JavaScript
- Frameworks
- DeepSeek Harness (DSH) · Cordis · zod · graphology · React Flow (@xyflow/react) · esbuild · Node.js
Tags
agent-preset · dsh-plugin · deepseek-harness · epistemic-loop · ontology · knowledge-graph · scientific-discovery · evidence-tracking · hypothesis-tracking · verification · local-first · research-assistant · memory · skills · nodejs · agent-framework