He Xiao
Community@hexiao-55 · HK
Wish you a good life!
Agent Skills by He Xiao
Showing 45 vetted skills indexed across 1 GitHub repositories.
taxonomy-build
Organize paper analysis results into a hierarchical taxonomy in taxonomy.md.
survey-pipeline
Automate end-to-end survey construction from topic refinement to final draft.
paper-write
Draft LaTeX papers section-by-section from PAPER_PLAN.md outlines.
paper-poster
Convert research papers into A0/A1 LaTeX conference posters.
survey-brainstorm
Refine vague research ideas into focused survey scopes through guided dialogue.
training-check
Monitor WandB metrics to detect NaN, divergence, or stalled GPU utilization.
result-to-claim
Evaluate experimental results against claims and route next steps.
research-refine-pipeline
Chain research refinement and experiment planning into a concrete proposal and roadmap.
survey-write
Generate a structured survey draft from taxonomy and gap analysis files.
auto-review-loop-llm
Automate multi-round research review cycles with an OpenAI-compatible LLM.
experiment-plan
Create claim-driven experiment plans with run order and compute budgets.
idea-creator
Generate and rank publishable research ideas from a broad direction.
paper-analysis
Analyze papers by topic, arXiv ID/URL, or PDF into 8-dimensional classification files.
grant-proposal
Draft a complete NSF CAREER grant proposal from research ideas and literature.
monitor-experiment
Monitor remote experiment runs via SSH and summarize JSON results.
research-pipeline
Automate research workflows from fuzzy ideas to submission-ready manuscripts.
research-refine
Transform vague research directions into anchored, implementable plans.
ablation-planner
Design structured ablation studies to validate paper claims.
idea-discovery
Orchestrate an end-to-end idea discovery pipeline from literature review to pilot planning.
paper-compile
Compile LaTeX sources into submission-ready PDFs with latexmk and auto-fixes.
paper-slides
Convert compiled research papers into Beamer PDF and PPTX slide decks with speaker notes and talk scripts.
paper-figure
Create publication-quality plots and LaTeX snippets from experiment results.
run-experiment
Automates deployment and execution of ML experiments on local or remote GPU servers.
auto-review-loop
Automate multi-round research reviews with iterative critique and fixes via Codex MCP.