Shared Sessions for Claude Code
링크로 Claude Code 세션을 공유하면 팀원이 각자 Claude Code에서 참여해 함께 작업합니다
Claude Code 플러그인 마켓플레이스 생산성과 학습 플러그인
Claude Code plugin marketplace - productivity and learning plugins
플러그인Python · JavaScriptGitHub @JoshuaOliphant직접 설치운영자 미확인
써 보셨어요? 한 번만 눌러 주세요. 로그인은 필요 없어요.
아직 감지된 업데이트가 없습니다. GitHub 릴리스가 나오면 여기에 보입니다.
개인 계정의 공개 저장소 소유자는 GitHub로 로그인해 운영자를 확인하고 소개를 직접 관리할 수 있습니다. 관리자 승인은 필요하지 않습니다.
로그인한 뒤 확인 파일을 만들어 이 저장소의 기본 브랜치에 30분 안에 커밋해 주세요. 다른 브랜치나 포크의 파일은 인정하지 않습니다. 파일을 올린 뒤 아래에서 확인하면 관리자 승인 없이 관리할 수 있습니다. 자격을 유지하려면 파일을 남겨 두세요.
로그인이 필요하면 로그인 화면으로 이동합니다. 로그인 후 이곳으로 돌아와 원하는 확인 버튼을 다시 눌러 주세요.
설치형 프로젝트는 배포 도메인 없이도 확인할 수 있습니다. 개인 저장소는 GitHub 로그인으로, 조직 저장소는 위의 저장소 확인 파일로 직접 확인해 주세요.
README 전체를 불러오는 중입니다. 아래는 저장된 발췌입니다.
A collection of Claude Code plugins for productivity, learning, and software-development
workflows.
Nine plugins, grouped by focus. Current versions live in
.claude-plugin/marketplace.json and each plugin's
plugin.json (the source of truth) — they are intentionally not duplicated here so this
list can't drift.README 전문 보기 ↗A collection of Claude Code plugins for productivity, learning, and software-development workflows.
# Add this marketplace
/plugin marketplace add joshuaoliphant/claude-plugins
# Install any plugin
/plugin install <plugin-name>@oliphant-plugins
Eight plugins, grouped by focus. Current versions live in
.claude-plugin/marketplace.json and each plugin's
plugin.json (the source of truth) — they are intentionally not duplicated here so this
list can't drift.
Create evidence-based spaced-repetition flashcards using cognitive-science principles (Andy Matuschak's five properties of effective prompts). Quality validation, knowledge-type workflows (factual, conceptual, procedural, salience), anti-pattern detection, and deck/card management via the Mochi API.
Requires: a Mochi account + MOCHI_API_KEY.
/plugin install mochi-creator@oliphant-plugins
export MOCHI_API_KEY="your_api_key_here"
Institutional memory across sessions and projects: capture solved problems, retrieve past solutions before you start, and graduate recurring lessons into living context (CLAUDE.md / AGENTS.md). Structured YAML-frontmatter files for fast grep-based retrieval.
Skills: compound-capture, compound-retrieve, compound-graduate.
Process information for real understanding — an antidote to the illusion of clarity. The
explain-back skill makes you explain from memory, grades you against the real source,
teaches struggle-then-teach per gap, and outputs Mochi cards plus a resumable session
record.
Turn a Claude project or a folder into a learning harness for one subject, adapted from learnharness. Lesson sessions probe your edge, plan a minimal path, and teach one checked reasoning step at a time; Practice sessions train retrieval with a hint ladder; Field sessions grade an artifact you produce against a rubric committed before you start. See plugins/learn-harness.
One software-engineering workflow made from Matt Pocock's skills, superpowers, pstack, Imbue's
blueprint, and the retired autonomous-sdlc and stick-shift plugins: specs with Given/When/Then
acceptance criteria, tracer-bullet issues in your tracker, builds that fit tests into the existing
suite, and review by subagents on every change. It also carries what used to be the review-diff
plugin (comment on a diff in your browser) and jev-lint (semantic lint for Python).
compost:setup configures a repo and offers to switch off the collections compost replaces.
See plugins/compost.
Build web apps where an AI agent dynamically generates the HTML UI, using hexagonal / ports-and-adapters architecture with HTMX for interactivity and MCP tools for data operations.
Stack: Claude Agent SDK + FastAPI + HTMX + Tailwind CSS. Requires: ANTHROPIC_API_KEY.
/plugin install hexagonal-agents@oliphant-plugins
export ANTHROPIC_API_KEY="your_key_here"
Install a local, Docker-free OTLP observability stack (OpenTelemetry → Vector → JSONL,
optionally VictoriaLogs + VictoriaMetrics) into a project and instrument the code — with a
scan-and-propose step for domain-appropriate spans and metrics — then query that telemetry
as a development feedback signal (observability-query skill, status.sh --json detection
contract, scripted end-to-end verify.sh).
One headless Claude per component, watching its telemetry on a local launchd schedule through the read-only MCP servers you already have (Splunk, Prometheus, Kubernetes, ...). Page-worthy findings raise a clickable macOS notification that opens a triage session; every verdict (real, noise, known) feeds a ledger that teaches the superfan what normal looks like. See plugins/superfan.
Each plugin owns its version in plugins/{name}/.claude-plugin/plugin.json; that version is
copied into .claude-plugin/marketplace.json at publication time. Run
python scripts/check_marketplace_versions.py to confirm they match. See
AGENTS.md for full development and versioning guidance.
Contributions are welcome — feel free to submit issues or pull requests.
MIT