$ show glebis/claude-skills
This is a monorepo of ~100 Claude Code skills — self-contained plugins that extend Claude Code's capabilities across personal automation, research, content creation, development tooling, and communication. Each skill is a directory with a SKILL.md (trigger definition), scripts/, references/, and assets/. The architecture is plugin-based: Claude Code matches user requests to skill triggers and executes the associated scripts. Skills range from simple CLI wrappers (e.g., YouTube transcript extraction) to multi-step pipelines (e.g., PII redaction with confide, release orchestration with cull-release, interview-driven skill design with skill-studio). The codebase demonstrates strong patterns: reversible PII redaction with typed placeholders, exact permutation statistics, weighted scoring with Pareto frontiers, and structured TDD loops. However, it also has technical debt: hardcoded user paths, some deprecated library usage, external API coupling, and a few design-spec-only skills. The dominant character is application (CLI tools and agent plugins), though several skills involve data analysis (health queries, experiment evaluation, evidence grading).
The LLM identified 5 files worth highlighting for reuse:
Core library for reversible PII redaction using typed placeholders (e.g., [PERSON_0], [EMAIL_1]) with map-based rehydration. Implements Span detection, green_sha256 hashing, and mangled-placeholder tolerance. Self-contained and could be extracted into any project needing auditable, reversible anonymization.
Deterministic scoring engine that computes weighted criterion scores (effectiveness, safety, convenience, cost penalty) and identifies Pareto-optimal alternatives. Supports budget filtering, sensitivity rankings, and tournament comparisons. Isolated domain logic with no external dependencies — reusable for any multi-criteria decision problem.
Library for exact circular-shift permutation tests, event difference tests, and break difference tests with Benjamini-Hochberg FDR correction. Implements fixed family size discipline and bootstrap confidence intervals. Standalone Python module reusable for any n-of-1 or time-series experiment analysis.
CLI tool for querying Apple Health SQLite exports with multiple output formats (markdown, JSON, FHIR, ASCII charts). Includes deduplication logic for Apple Watch vs iPhone double-counting and topic-specific filters (sleep, activity, vitals). Self-contained script that could be adapted for any health data export.
Orchestrator for publishing markdown drafts to Telegram channels with frontmatter parsing, markdown-to-HTML conversion, media attachment, footer injection, and post-publish bookkeeping. Reusable as a library for any Telegram content publishing pipeline.