SKILL.md packages that extend Claude Code, Cursor, Copilot, and other AI agents.
Tags

nub
Compress and explore very large files or folders by adjusting level-of-detail: overview, index, or focused reading for codebases, logs, and large documents.

drift
Group and deduplicate multi-source events/articles using ID- and content-based grouping, reputation scoring, and canonical selection to pick the best version.

K.I.T (Known Issue Triager)
Consolidate audit reports into a canonical known-issues.json register, extend existing registers, or check whether a reported issue is already known.

obsidian-master-kit
Semantic vault organizer for Obsidian (pt-BR): scans note embeddings, detects duplicate/conceptually-similar notes, finds clusters missing MOCs, and proposes no

digital-marketing-pro
Universal import pipeline to load CSV, JSON, or Google Sheets into CRMs and marketing platforms with field mapping, validation, deduplication, consent checks, a

ds-skills
Post-process entity-match predictions to enforce symmetry and transitivity (graph closure) to improve recall in deduplication and record linkage.

ai-runbooks
Find potentially duplicate or similar cases before deep analysis by running a similarity check against existing cases and returning matching case IDs.

Claude Code Skills
Merges and deduplicates evaluation worker results into a single verified review summary for coordinators, applying an architecture gate and verification checks.

workflows
Search academic literature across multiple sources, deduplicate results, resolve DOIs, and surface trusted papers with concise takeaways.

literature-harvest
Reproducible local workflow to search PubMed/Crossref/OpenAlex, build candidate tables, download legally accessible PDFs/HTML, and deduplicate harvested literat

ds-skills
Generate geographically proximate candidate pairs for entity matching using haversine-distance KNN, with optional partitioning by country to improve recall and