Productivity plugins

AI job search pipeline — evaluate offers, tailor CVs, scan portals, track applications

5 skills

Reduce, redact, preserve, and retrieve large AI-agent context through XContext.

1 skill · 3 commands · 1 MCP server

Persistent memory for Claude Code - never forget what you learn

3 skills · 1 agent · 1 MCP server

Stay fit while Claude works. Random calisthenics in your status line, delivered by an unhinged rotating cast of fitness personalities powered by Haiku.

4 skills · hooks

Shared development standards including makefile-standard, dockerfile-standard, compose-standard, ci-standard, and go-project-standard

5 skills

Codex-native DRI orchestration, five role agents, durable Beads mail, and agent-teams setup.

4 skills · hooks

Fraud verdicts, route checks and caller-ID verification for every SIP call, operated by Claude.

4 skills · 1 MCP server

Ancher — personal knowledge base & research assistant. Skills + hosted MCP server.

4 skills · 1 MCP server

Multi-agent orchestration (Planner / Reviewer / Executor / Verifier) with human checkpoints.

1 skill · 4 agents

Point-and-click annotation editor for AI artifacts. 3-5 second feedback loops. Beautiful HTML rendering.

2 skills · 2 commands · hooks

Offload boilerplate, summarization, and simple analysis tasks to a local or remote LLM to reduce Claude token usage

5 skills

Playwright-powered browser automation with persistent Chrome profile, visible browser, and fast screenshot/click/scroll/type operations

3 skills · 1 agent · 1 MCP server

Webhook in, agent out. Skills for installing and exposing skillhook, authoring webhook skills and operating Skillhook Cloud, plus two MCP servers: skillhook (status, skills, secrets, jobs, Tailscale exposure, signed test webhooks) and skillhook-cloud (every machine of the organisation, as the dashboard shows and runs it).

3 skills · 2 MCP servers

Daily interview-driven PMF logs, a hypothesis/evidence ledger, weekly reviews, a generated index, and a one-time importer — with non-negotiable no-fabrication rules. Drives a founder's customer-discovery / product-market-fit process entirely in indexed Markdown.

5 skills

AI course design, slide PDF generation, narrated video production, and course video publishing skills.

5 skills

Skills for working with Steady (runsteady.com) from AI-powered tools — pull your current work context before related tasks, write code against the Steady v2 API, drive the steady CLI, and compose clear check-ins and goal updates.

4 skills · hooks

Turn your own Claude Code history into a personalized, self-updating coding-agent skill. Builds an evidence-grounded persona, generates a 'working-with-you' skill tuned to keep you in flow, and keeps learning from your newest sessions.

1 skill · 3 commands · hooks

Transcribe recordings, extract patterns, track progress, and gain insights from therapy sessions and journals

5 skills

/memarium digests Claude Code, Copilot Chat, and Codex sessions into typed memory (episodic / semantic / procedural / core) + an entity wiki + distilled Q&A. A SessionStart hook auto-injects the project primer; /memarium-context and /memarium-recall provide deeper retrieval.

4 skills · hooks

Utility skills for daily work

5 skills

Mediates interaction with the Atlassian MCP server to minimize context bloat. Provides focused Jira and Confluence access with digest-style output by default.

1 skill · 2 commands · 1 agent · hooks

Bidirectional Telegram messaging for Claude Code. Send messages, images, and receive commands remotely. Includes auto-enter and remote permission control via Telegram.

4 skills · 1 MCP server

業務フロー系スキル（要求文書・仕様書の生成と監査、文書の言語化レビューと改稿、Codex 向け Workflow 内部互換層）

5 skills

Full lifecycle manager and optimizer for agentic skills — author, densify into micro-kernels, pack, unpack, pin, sym, diff, snapshot, bump, archive, fork, publish, and git-guard.

5 skills

Knowledge and interface design skills.

5 skills

Documentation authoring workflows: co-authoring docs, strategic design documents, and polished editable SVG diagrams.

4 skills · 1 agent

Push notifications via Pushover when Claude needs your attention

1 skill · 3 commands · hooks

Claude Code hooks that make Claude finish the job: a decision model on OpenRouter judges each tool call and final report, and sends Claude back to work on script edits, unrequested config relaxation, building before searching, and unfinished, verbose or unmeasured reports

4 skills · hooks

Read KOSIS (Statistics Korea) national statistics

5 skills

Тест «ИИ-мастер»: замер навыков работы с ИИ по шкале из исследования Anthropic — 11 привычек по вашей реальной истории диалогов или экспресс-тест, блок неиспользуемых фишек Claude Code, страница результата, карточка для сторис и персональный тренер. Всё считается локально.

4 skills · 1 command