By dadwadw233
Team collaboration protocol skills for handoff, checkpoint, team progress, and stale Obsidian docs refresh workflows.
Use when the user invokes `$checkpoint`, asks for a checkpoint, or asks to update CURRENT/NEXT/RISKS/TODO without committing or pushing.
Use when the user invokes `$docs-refresh`, asks to update Obsidian project docs from a stale audit/dev record, or wants old docs archived while active docs become current.
Use when the user invokes `$handoff`, asks for a handoff, or asks to finish a team-collab session by updating CURRENT/NEXT/RISKS/TODO and syncing docs.
Use when the current repo is explicitly governed by team-collab: `obsidian-docs/` exists in cwd/ancestor, repo `AGENTS.md`/`CLAUDE.md` imports team-collab, the cwd is under a registered code/docs path, or the user asks for team-collab handoff/checkpoint/docs governance/Feishu automation. Do not use solely because a global config file exists.
Use when the user invokes `$team-progress`, asks to sync recent progress from other team members, or asks who has blockers or PR/MR items needing review in a team-collab project.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Keep your AI coding agents on the same page — a shared, file-based way to track project state, hand off work, and let several agents collaborate. Installable for Claude Code, Codex, Cursor, VS Code, Cline, OpenCode, Continue, Gemini CLI, and manual workflows.
Quick Start · Adapter Matrix · Multi-Agent Collaboration · 中文简介
When a team leans on AI coding agents, work slips through the cracks between sessions: nobody is sure what state the project is in, who owns which TODO, why a past decision was made, or what the last session actually did. Handoffs turn into scrolling back through old chat logs.
team-collab-skills gives your agents a shared, file-based routine so any agent — or teammate — can pick up where the last one left off:
CURRENT.md (where things stand), NEXT.md (what's next), RISKS.md, and TODO.md. Everything important lives in the repo, not in a chat window.$checkpoint to save progress mid-task, $handoff to wrap up a session, $team-progress to see what teammates did recently, and $docs-refresh to bring stale docs back up to date.@owner so two agents don't grab the same task at once.This repository is the source of truth that agents actually load at runtime: the skills, protocol references, adapters, and manifests. A companion command-line tool, @embodot/collab, is published on npm and adds an installer, health checks, and the multi-agent commands.
claude plugin marketplace add dadwadw233/team-collab-skills
claude plugin install team-collab@team-collab-skills
Claude loads the protocol automatically when it sees a clear team-project signal — for example an obsidian-docs/ folder, a project AGENTS.md, references to CURRENT/NEXT/RISKS/TODO, or a direct request like "do a handoff".
codex plugin marketplace add https://github.com/dadwadw233/team-collab-skills.git
Then use the entrypoints:
$checkpoint
$handoff <topic>
$team-progress 24h
$docs-refresh <audit-doc-or-topic>
If Codex passes /handoff, /checkpoint, /team-progress, or /docs-refresh through as plain text, treat it the same as the $... form.
If you also use the @embodot/collab command-line tool, it can install these skills for every agent at once and check your setup:
npm install -g @embodot/collab@latest
team-collab install-skills --agent all --force
team-collab doctor --project <project>
Adapter files are deliberately thin pointers back to the protocol — don't copy the full protocol into each tool's rule file. The source of truth is skills/protocol/.
npx claudepluginhub dadwadw233/team-collab-skills --plugin team-collabAnalyze your AI conversation history and generate a beautiful developer personality portrait with MBTI type, famous person match, and developer rating.
A single-skill package for generating harness blueprints for agentic systems.
A single-skill package for designing useful, controllable, extensible agentic systems.
Security guard for AI coding agents - detects secrets, blocks dangerous operations, provides a security dashboard
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
Build and maintain an LLM-curated personal knowledge base in your project — Andrej Karpathy's LLM Wiki pattern, designed to scale to thousands of pages without becoming a context bottleneck. Now with an optional compiled graph layer for typed, provenance-backed relationships.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Harness-native ECC plugin for engineering teams - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.