sickn33/agentic-awesome-skills

AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 1,987+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.

What it solves

AI coding agents often struggle to select, validate, and execute complex workflows without leaving behind an opaque or unreviewable trail of actions. AAS Core provides a structured way for agents to search a local catalog, compose specific skill stacks, and generate immutable, reviewable plans before any changes are applied to a project.

How it works

The system uses a Model Context Protocol (MCP) to give agents like Claude or Codex read-only access to a local catalog of nearly 2,000 skills. The agent inspects the project, searches for the necessary capabilities, and selects specific skill IDs. AAS Core then validates these selections in memory. The user can then use a CLI to generate an immutable plan or persist the selection as a JSON manifest for human review.

Who it’s for

It is designed for developers using AI-powered coding tools (such as Claude Code, Cursor, or Codex CLI) who want to transform unpredictable agent behaviors into reproducible, inspectable, and validated workflows.

Highlights

  • Agent-first composition: Allows agents to search and select exact skill IDs from a local catalog.
  • Immutable planning: Generates per-target plans that can be reviewed before any files are modified.
  • Extensive skill library: Includes over 1,900 skills covering development, security, DevOps, and data analytics.
  • Multi-tool compatibility: Supports various environments including Cursor, Claude Code, Gemini CLI, and GitHub Copilot.
  • Reviewable state: Uses JSON manifests to preserve agent-selected stacks and execution evidence.