skills

Name: skills
Rating: 69 (1 reviews)
Author: ZibbyDev

MCP Server

ZibbyDev/skills

@zibby/skills is a companion package for @zibby/agent-workflow that provides built-in and custom tool/MCP skill definitions for AI agent nodes.

7 0MITUpdated 2026-08-13

Good

View on GitHub Open source file

Overview

This package acts as the skill layer for graph-based AI agent workflows. It features a unified skill factory that supports both custom function-based tools and Model Context Protocol (MCP) server wrappers, making it easy to connect tools like browsers, issue trackers, and databases to agents running via Claude SDK or Cursor CLI.

Capabilities

▸Unified skill creation and auto-registration via the skill() factory
▸Function skills with custom input validation and async handlers
▸MCP skill integration with dynamic environment variable resolution and command fallback
▸Multi-agent support wiring skills into Claude SDK and Cursor CLI configurations

Best for

Defining custom programmatic tools with input schemas and asynchronous handlers for AI agents, Wrapping existing Model Context Protocol (MCP) servers for workflow nodes, Integrating external services like Jira, GitHub, Slack, and Playwright browsers into agent workflows

Works with

Claude CodeCursor

Evoa Score breakdown= Σ (score × weight)

Task usefulness20%77 → +15.4

実タスクにどれだけ役立つか（機能の豊富さ・用途の明確さ）。 — AIによるcapabilities/use-cases解析

Code quality15%90 → +13.5

実装・指示の品質。 — AIによるSKILL.md/README解析

Maintenance15%100 → +15.0

リポジトリがどれだけ活発に保守されているか。 — GitHub 最終push日時の新しさ

Documentation12%37 → +4.4

ドキュメントの充実度・分かりやすさ。 — README/独自要約の情報量

Security15%50 → +7.5

危険・不審な挙動が無いか。 — AIによるセキュリティレビュー

Originality10%85 → +8.5

ありふれたラッパーではない独自性。 — AIによる独自性判定

Popularity8%23 → +1.8

コミュニティの採用度。 — GitHub Stars/Forks（対数スケール）

Compatibility5%50 → +2.5

対応AIエージェントの広さ。 — AIによる対応エージェント判定

Weighted total68.7 / 100

ライセンス不明/制限あり（Red）のSkillは総合スコアに0.85倍の補正を適用します。ランキングはこのScoreのみで決まり、広告で変わりません。算出方法の詳細 →

Security considerations

MCP skills and function handlers may execute local processes, make outbound network requests, or interact with sensitive external APIs requiring secure handling of environment variables and API keys.