Evidence-grounded AI agent for Java code repair with LLM-guided patches and human-in-the-loop review.
Evidence-grounded research agent for Claude Code & Codex: every claim gets a confidence rating + source-family tag, and adversarial gates block unsupported claims before they reach your report. Ships with a multi-model dispatcher.
Evidence-grounded AI research workflow for traceable claims, explicit uncertainty, causal-language checks, and reviewable JSON, Markdown, and Excel exports.
Local-first agentic research platform that turns project bundles into evidence-bound scientific papers with deterministic gates, scientist-panel review, fault localization, and rollback.
面向 macOS 的本地优先论文研究助手，支持原文证据定位、页级引用、长任务恢复与按需公式 OCR。
AI-powered talent sourcing agent — multi-source discovery, evidence-grounded scoring, pluggable adapters
Calculate the real social distance between two public X/LinkedIn profiles — under strict bidirectional-interaction evidence, with an audit trail.
Evidence-grounded AI paper reading: local PDF or Zotero to traceable Markdown/Obsidian notes with verified pages, key figures, and fidelity checks.
AI specialists tailored to your project. They catch themselves when wrong. Audit-grade by default agentic harness.
GenAI decision-support app for Washington State Adult Family Home pre-admission intake: evidence-grounded care plan, CARE acuity factors, and capability-gap analysis. Streamlit + Claude, with a deterministic interview and a baseline-compared evaluation.
Evidence-grounded scientific agents with replayable evaluation and human review — the public integration + evaluation layer for AI-for-science workflows.
Evidence-grounded RAG platform combining hybrid retrieval, reranking, citation verification, and fail-closed answer generation.
D.A.R.I.A. — a domain-specific agentic RAG research assistant for evidence-grounded answers from a curated research-paper knowledge base, built with LangGraph, memory, evidence curation, critic-based validation, and corrective retrieval.
Evidence-grounded local RAG system for academic paper analysis on consumer hardware.
To associate your repository with the evidence-grounded topic, visit your repo's landing page and select "manage topics."