jluo41

by Junjie Luo

Single-entry Obsidian intake and chronicle tools. Use /note to capture readable daily logs, people notes, project/matter notes, tasks, timelines, meetings, emails, and raw material.

by Junjie Luo

Skills for the haipipe data + NN + endpoint pipeline. Tier-1 user umbrellas (/haipipe-data, /haipipe-nn, /haipipe-end, /haipipe-project, /haipipe-individual) parse intent and dispatch to per-stage / per-target / per-risk-profile Tier-2 specialists. Covers stages 0-2 (per-individual data contract), 1-4 (project-wide data build), 5 (model training), 6 (endpoint packaging + deploy targets: SageMaker, Databricks, Flask, MLflow), and per-project scaffolding.

by Junjie Luo

Normalizer utilities for the haipipe pipeline: free text in a cohort's own dialect, resolved against a reference bank into numbers that carry their own provenance. One skill per noun — describe-food resolves a FoodName to USDA nutrition; describe-exercise resolves an ExerciseType to a PA Compendium MET and, when the log stated minutes and the body mass is known, an estimated kcal — and haipipe-norm owns the contract all of them obey.

by lewislulu (upstream)

HTML PPT Studio — author professional static HTML presentations (37 themes (academic-report = house default), 31 layouts, 15 full-deck templates, 27 CSS animations, 20 canvas FX, presenter mode with speaker notes). Includes html-to-svg (per-slide 1280x720 SVG export for PowerPoint import). Vendored from https://github.com/lewislulu/html-ppt-skill.

by Junjie Luo

In-lab human interaction with deployed prediction endpoints — the Claude Code chat is the UI. CONSOLE mode (default): patient-first on-demand inference — list patients, view the full chart, list available models (Endpoint_Set packages + live status), and run 'this model on this patient': the endpoint-predict MCP tool prepares the payload (required tables + trigger record), POSTs to the endpoint, and returns the score verbatim with a data-gaps report. STUDY mode: formal reader protocol (frozen review_bundle.json, blind→assisted session, decision-influence metrics vs gold). Endpoint-agnostic: local Flask, Databricks, SageMaker — same wire contract.

by Junjie Luo

JHU Research-IT helpdesk chatbot. Answers how-to questions about Johns Hopkins research IT infrastructure (PMAP, SAFE, SAFER, Discovery, Databricks, Crunchr, Epic Cosmos, REACH) by reading a local WIKI/ knowledge base. Invoke as /jhu-research-it.

by Junjie Luo

Human-grounded subjective labeling in two sibling sides: Label Building calibrates and freezes one identified human's construct into a signed Label Handoff; Label Scanning qualifies executors on sealed human gold, labels the corpus under that frozen handoff, and closes with an independent probability audit.