Claude Code and Codex plugins their authors filed under "productivity". All categories.
A verified header line at the top of every Claude reply (time, date, branch...) and a configurable status-line footer (session tokens, plan limits, clock...), both from one config file. Python standard library only.
1 skill · hooks
API gateway for calling third-party APIs with managed auth.
1 skill
A box above the prompt that says what Claude is doing right now in tail-claude-hud's icons and colors, and expands to show what it has done.
Claude Code software delivery workflows from this repo
9 skills · 11 agents · hooks
Run, validate, and analyse SUEWS urban climate simulations from your AI assistant. Bundles workflow rules, the `suews-mcp` Model Context Protocol server (validation, inspection, summarise/compare/diagnose, init/convert, versioned knowledge pack), fresh-site simulation guidance with assumption ledgers/readiness levels, forcing-data checks, output interpretation, and common-error diagnostics.
13 skills · 1 MCP server
Long-term memory for Claude Code — slash commands, skills, opt-in session hooks, codebase indexing, and a deep-retrieval subagent.
4 skills · 6 commands · 1 agent · 1 MCP server · hooks
Plan-first development workflow. 3-layer enforcement, .status-based plan tracking, auto quality checks.
8 skills · 2 agents · hooks
Track 80+ AI entities across Reddit, HN, GitHub, HuggingFace, arXiv, and agent X/Twitter search. Daily-scheduled with scored, deduplicated Markdown reports. Supports unattended cron execution.
8 skills · hooks
Pomodoro timer skill that tracks and learns from your focus sessions.
1 skill
Shows above the prompt which other Claude Code sessions are waiting for you, and reports this session to the local hopper-hq daemon.
hooks
Refuses force-pushes, pushes to protected branches, git commands that skip hooks and kills by process name; optionally bans folders, keeps subagents out of shared checkouts and stops whole-file reads of big files.
Refuses a subagent spawn that names a pinned model id, and enforces your own per-agent-type model rules.
Keeps a registry of interactive sessions so /resume-sessions can list and reopen the ones a crash, reboot or closed terminal interrupted. Windows, PowerShell, WezTerm or Windows Terminal.
Give Claude Code a private memory that lives on your own machine — remembers across sessions, works fully offline, no cloud. 100+ MCP tools: fast hybrid search, automatic chat capture, contradiction tracking, and one-command GDPR export/delete.
16 skills · 15 commands · 1 MCP server
BATHOS — 17역할 7웨이브 AI Workflow Agent 메서드 패키지 (Codex CLI 네이티브 플러그인: skills + hooks)
31 skills · hooks
长篇网文创作系统（skills + agents + data chain + RAG）
8 skills · 20 agents
Nova the Optimal AI Free: one vivid Nova front door, ambient MIND, governed continuity, projects, personal knowledge, and twenty-seven native practical capabilities.
28 skills
A curated set of skills for each stage of development — propose, spec, design, plan, implement, ship.
27 skills
Give your AI coding agent a permanent second brain. Connect Claude to an Obsidian vault to auto-record decisions and recall context.
5 skills · 8 commands · 5 agents · hooks
Utilities - integrations, dry-run, prototyping, asset management
17 skills · hooks
Fifteen evidence-first Codex workflows, one legacy-shim lifecycle manager, fifteen specialist role cards, shared gates, and reviewable artifacts.
16 skills · hooks
Guardrail hooks, a plan → implement → verify → crosscheck loop, a /critical-review exit gate with a pluggable reviewer backend, doc-tiering commands, a /setup-check, a browser-MCP skill and a terse output style for Claude Code.
1 skill · 10 commands · hooks
bridge_CC-Codex: guided setup plus bounded implementation, advice, and review across Claude Code and Codex.
7 skills
Static Python structure for Claude Code: indexing, queries, test impact, reference renames, integration, and session telemetry debriefs
6 skills
Сжимай ответы Claude на русском или казахском. -65-80% токенов. Техника целиком. /cavemanov или /cavemanov-kk.
4 skills · hooks
Author, validate, and blockwise-edit GEML documents, and navigate a project's call graph as GEML codemaps: two skills plus the geml MCP server (runs via npx @geml/geml, no separate install).
2 skills · 1 MCP server · hooks
Ultra-compressed communication mode.
2 skills · hooks
Build data-connected dashboards, internal tools and back-office apps in ToolJet, then refine them visually. Requires a ToolJet connection.
2 skills · 1 MCP server
Laziest solution that actually works.
1 skill · hooks
HandoffKit by DynG AI brings the PlatformPilot.ai coordination model to Codex: explicit agent handoffs, actor-style mailboxes, routed ownership transfer, pub/sub awareness, fan-in joins, budgets, dead letters, and traceable multi-agent workflows.
2 skills