Agent Skills by Chan Uni
Showing 32 vetted skills indexed across 1 GitHub repositories.
research
Orchestrate an end-to-end research workflow with resumable YAML-defined pipeline stages.
daily-arxiv
Discover arXiv papers and ingest relevant ones into a research wiki.
/init
Build an EmpiricalWiki workspace from PDFs and notes with parallel ingestion.
/variable-map
Consolidates empirical variables from wiki pages into an executable dictionary with comparison table.
exp-design
Generate claim-driven experimental plans with baseline, validation, ablation, and robustness blocks.
/paper-draft
Generate a complete LaTeX paper draft from a wiki-linked PAPER_PLAN.
discover
Generate ranked paper shortlists from anchor, topic, or wiki seeds.
ask
Retrieves wiki pages and synthesizes evidence-grounded answers with citations.
/prefill
Generate slugged wiki foundation pages from Wikipedia or LLM definitions.
ingest
Convert research papers into cross-referenced EmpiricalWiki pages with bidirectional links.
reset
Reset EmpiricalWiki workspaces to a clean scaffold by scope.
setup
Configure missing environment variables for EmpiricalWiki external services.
/exp-status
Monitor experiment runtime states and flag anomalies from screen or GPU/SSH runs.
/stata-plan
Generate a Stata execution plan from EmpiricalWiki design pages.
edit
Edit raw source files and wiki pages while maintaining reverse links.
paper-compile
Compile LaTeX papers to PDF with diagnostics and compliance checks.
/empirical-ingest
Extract variables, datasets, and models from empirical papers into wiki pages.
/novelty
Cross-check research ideas against web, Semantic Scholar, and EmpiricalWiki sources.
/review
Review research artifacts and produce structured scores, weaknesses, and improvement plans.
ideate
Generate ranked empirical research ideas from the EmpiricalWiki knowledge graph.
/paper-plan
Generate venue-aware paper outlines from wiki claim graphs.
empirical-design
Generate an empirical research design from wiki pages with citations.
/refine
Iteratively review and edit wiki-linked research artifacts until target quality scores are met.
exp-eval
Judge completed experiments and update wiki claims with verdicts.