Your agent can do more. Find the skill that makes it happen.
Find your next superpower If you are an agent, refer to our llms.txt for full access."..." for exact phrases · Try typography, debugging or spreadsheetDefines experiment protocols and evidence structures covering datasets, baselines, metrics, ablations, and result-table contents. Preserves real values while leaving visual table rendering and broad literature retrieval to specialist skills.
Provides the initial preflight for CCFA research, review, retrieval, experiments, visuals, and maintenance. Encourages direct communication, removes empty defensive framing, and preserves evidence and uncertainty while limiting prose edits to the authorized task.
Refines early CCF research ideas and explores possible directions or recovery paths. Produces concrete problems, insights, mechanisms, and evidence plans; idea evaluation and manuscript editing are separate tasks.
Reviews, scores, or ranks research concepts, including ideas extracted from manuscripts. Defaults to concept-only assessment without evaluating experiments, and does not develop ideas or review manuscript evidence.
Audits existing claims, numerical consistency, terminology, citations, and BibTeX context against supplied or verified evidence. Its scope excludes full scientific review and discovery of new literature.
Scans arXiv, OpenReview, venue feeds, labs, and competitors for recent work that overlaps with a research project. Provides dated evidence and changes since the previous scan for recurring novelty monitoring.
Retrieves and verifies external literature, prior art, datasets, benchmarks, and potential citations. Supports related-work searches and research opportunity maps, with recurring monitoring and existing-citation audits handled separately.
Produces structured scientific or writing reviews of manuscripts, including weaknesses, support for conclusions, submission maturity, and version comparisons. Reviews can include scores, but prose rewriting and concept-only idea evaluation are separate tasks.
Distills supplied paper PDFs into reusable writing exemplar cards and registers a personal library when requested. Preserves source attribution and transfers writing logic rather than reconstructing figures, drafting manuscripts, or reviewing papers.
Writes and edits abstracts, sections, slides, and other CCF manuscript content, including revisions based on reviews. Preserves supplied evidence and formatting; assessment without rewriting and rebuttal writing are separate tasks.
Breaks down tasks and coordinates research workflows, project status, goals, gates, artifacts, and ccfa.yaml state. Supports explicitly requested end-to-end coordination while specialist skills produce research outputs and scaffolding.
Creates reproducible CCF project workspaces with folders, LaTeX templates, artifact directories, and ccfa.yaml. Preserves existing files and leaves workflow planning and research content to other skills.
A growing collection of real, public skills. Descriptions and instructions indexed 25 Sept 2026.
Skills give your agent reusable instructions for a specific job. Pick one, read what it does, and bring it into your workflow.
A name only tells half the story. Search the full description and instructions to find the right fit.
Read the skill, visit its source, and see exactly what you’re adding to your agent.
Copy the install command from a skill page and run it in your project.
npx skillycli add owner/repo --skill name