Claude Code and Codex plugins their authors filed under "productivity". All categories.
Scrolls: a git-native markdown system for agentic memory that persists project state across sessions and agents (/scrolls-setup, /scrolls-update, /scrolls-hide, /scrolls-unhide, /scrolls-remove, /scrolls-restore, /scrolls-help, /scrolls-explainer).
Repo-related skills: acting on a repository's own hosted state (issues, pull requests, releases, and similar). Files GitHub issue(s) for a PR's work and keeps the PR's title/description in sync with them.
Animated architecture and network diagrams with flowing packets, orthogonal routing, selectable nodes, and scenario playback.
Claude Code mod: live band and pane for task progress, subagents, Downshift Compactor figures and routing decisions. Observes only; the downshift binary is optional.
125 battle-tested skills, agents, commands, and hooks for TDD, security scanning, code review, and autonomous development.
145 skills · 77 commands · 36 agents · 5 MCP servers · hooks
ITSOL engineering skills and subagents with compact progressive-disclosure routing, capability profiles, governed/autonomous-planned/direct workflows, repo memory, current technology research, TDD, security, infrastructure, review, OpenCode, and Pi
116 skills · 113 agents · hooks
Unified Project Manager. PM automation + 100 domain expert harnesses + goal-backward validation.
200 skills
226 context-engineered persona skills from PromptLibrary 3.0 with self-refine loops, quality dimensions, and cognitive scaffolding
200 skills
The required bstack companion skills (references/companion-skills.yaml, required: true) plus the bstack primer, installed from the broomva/skills monorepo. Pair with the bstack plugin; for every other skill add the broomva/skills marketplace.
117 skills
Operational toolkit for freelance and fractional CTOs — contracts, project-takeover SOPs, process-engineering + requirements-first pipelines, ops playbook/runbook authoring, clean-code standards (DSMS/IFS/CPS/PRDS/BrS/QT4L), and multi-harness PRDS pre-push hooks.
100 skills · 2 commands · hooks
AI agents modeled after a software startup org chart — specialized roles that synthesize to discover, build, ship, operate, and learn from production-quality software.
47 skills · 22 commands · 25 agents · hooks
Agentic Engineering toolkit — 42 skills, 13 agent personas, 36 slash commands. Full UI chain (brand → design-system → components → pages → dataviz → polish) plus motion-system reference. forge:meta uses generated_from hash snapshots so upstream edits cascade zero downstream writes. /sync + /forge-migrate + /feedback + bin/forge-check.sh. TDD-tested skills.
42 skills · 36 commands · 13 agents · hooks
ARIS research workflow as a Claude Code plugin: literature review, idea discovery, experiments, paper writing, rebuttal, slides — with a cross-model Codex reviewer. Run /aris:setup once after installing.
89 skills · 1 command
Cowork-optimized slim build of Artibot — marketing, content, data, design, and CRO/SEO skills for knowledge workers using Claude Cowork. No hooks, no developer-only components.
46 skills · 21 commands · 12 agents · 1 MCP server
Guided instructional design toolkit — design cmi5-compliant courses and 1-on-1 session plans using CONTEXT→CONCEPT→BUILD→SHIP→REFLECT formula, Builder's Bloom's progression, Ship-First Design, and Kirkpatrick L1-L4 evaluation. Produces JSON + Markdown + HTML visualization + Marp slide decks.
22 skills · 37 commands · 20 agents · hooks
Balsm shared skills, commands, and agents — install once, available in every project.
66 skills · 1 command · 1 agent · 3 MCP servers · hooks
Data Engineering Knowledge Capture. The data-plane second brain (29 nouns). Data jobs are IngestionJob, not AGER Workflow. Fail-closed 1/4-window ContextPack token budget.
24 skills · 21 commands · 25 agents · hooks
HITL AI-Driven Development (2.x line, current). 50 skills, 6 agents, enforcement hooks, and templates for the full governed delivery workflow — the step plan is derived per change from impact analysis — plus the platform readiness workflow (onboarded to delivery-ready) with a hard production-deploy gate. Start work with /hitl:dev-start-change.
60 skills · 5 commands · 6 agents
AI coding governance OS. Provides 4-Gate approval system, Intent/Mode classification, Skill Policy Router, Evidence Ledger, and agent control layer for Claude Code.
42 skills · 6 commands · 18 agents
HITL AI-Driven Development — legacy 1.x line (v1.1.0). The numbered 31-step delivery workflow plus the platform readiness workflow and hard production-deploy gate. For teams not yet ready to move to 2.x; feature-frozen except critical fixes. Install this OR hitl, not both. Start work with /hitl:dev-start-change.
50 skills · 5 commands · 6 agents
System Architecture Capture. The what-is-running second brain (139 architecture/runtime nouns). Fail-closed 1/4-window ContextPack token budget.
21 skills · 18 commands · 20 agents · hooks
Claude communication style: critical peer, concise output, confidence levels, research-first approach
46 skills · 7 commands · 2 agents · hooks
Session and task management: continuity, verification, design exploration, git worktrees, parallel agents
46 skills · 7 commands · 2 agents · hooks
Project Knowledge Capture. The why second brain: meetings, decisions, TicketLink / WikiTicket work types. Fail-closed 1/4-window ContextPack token budget.
28 skills · 25 commands · 2 agents · hooks
Cortex. A user-owned Markdown second brain with shared deterministic memory utilities. Works with Claude Code and Cowork, and ships portable adapters for Codex and ChatGPT Work without forking the canonical workflows.
27 skills · 24 commands · 3 agents · hooks
Curated portable skills (curated-skills.txt): workflows, composites, and operator utilities.
51 skills
Founder-class personas (Steve Jobs, Elon Musk, Marcus Aurelius, Jensen Huang, Margaret Hamilton, Phil Jackson, Sara Blakely, Maya Angelou, Jony Ive, Rick Rubin) — strategic decision-making and orchestration.
23 skills · 22 agents · hooks
SDLC skills for MPT extensions: Jira, branching, commits, pull requests, reviews, documentation, and releases.
33 skills · 11 commands · hooks
Lightweight SDLC workflow: capture backlog items, capabilities, decisions, drivers, milestones, principles, products, references, standards, tasks, and terms as markdown files with validated frontmatter. Ships task-level skills (task-work, task-review, task-ensure-ready, task-define) plus a JSON-schema-driven frontmatter validator every entity type plugs into.
41 skills
25 skills, 5 rule reference files, 2 subagents, and 4 deterministic hooks for solo founders and designer-developers.
36 skills · 2 agents · hooks