Loading...
Loading...
Found 163 Skills
Use when drafting, revising, structuring, reviewing, polishing, or preparing academic manuscripts, proposals, related work, LaTeX papers, abstracts, introductions, methods, limitations, or submission-ready scholarly text.
Use when migrating messy academic research repositories, downloaded archives, proposal folders, ad hoc notebooks, scripts, datasets, or paper assets into the standard research project structure.
Use when planning, running, auditing, or documenting systematic reviews, scoping reviews, PRISMA-style flows, screening decisions, inclusion criteria, exclusion criteria, or reproducible literature searches.
Use when developing academic research ideas, framing research questions, defining scope, novelty, contribution claims, hypotheses, operational definitions, evidence plans, or publishable positioning before SOTA, experiments, or writing.
Use when choosing, comparing, or preparing for computer science venues, conferences, workshops, journals, tracks, deadlines, reviewer expectations, paper fit, or publication positioning.
Use when creating, repairing, refactoring, validating, or documenting an academic research repository structure, including wiki, sources, SOTA, outputs, agent docs, tests, and reproducibility folders.
Use when responding to academic reviewers, planning revisions, writing rebuttals, mapping reviewer concerns, deciding concede/defend/reframe actions, or preparing camera-ready changes.
Use when building a state-of-the-art review, literature matrix, related work section, systematic search, paper comparison, gap analysis, or citation-backed synthesis.
Use when building research dashboards, annotation tools, data browsers, paper-supporting demos, SOTA explorers, experiment viewers, or frontend interfaces for academic projects.
Paper Analyst — Responsible for in-depth paper reading, extracting method details, and building comparison tables. Activated when assigned to analyze papers by research supervisors or literature investigators. Conduct structured analysis on the Top 20 core papers, generate paper analysis cards and cross-paper comparison tables.
Multi-source search and deduplication layer with intent-aware scoring. Integrates Brave Search (web_search), Exa, Tavily, and Grok to provide high-coverage, high-quality results. Automatically classifies query intent and adjusts search strategy, scoring weights, and result synthesis accordingly. Activated for "deep search", "multi-source search", or when high-quality research is needed.
Submit or run an ML experiment on a compute environment (local, SLURM HPC, RunAI/Kubernetes). Use when the user wants to launch a training run, submit a job, run ablations, or execute an experiment script on any compute cluster.