agent-skills-cli and speckit-agent-skills
The first tool, a universal CLI for agent skills, acts as an ecosystem sibling to the second tool, which provides agent skills for a specific framework (Spec Kit), as the former could potentially manage and sync skills including those offered by the latter.
About agent-skills-cli
Karanjot786/agent-skills-cli
Universal CLI for Agent Skills. Access 40,000+ skills from SkillsMP and sync them to Cursor, Claude Code, GitHub Copilot, OpenAI Codex, and Antigravity.
Provides token-aware context budgeting, conflict detection, and quality scoring (0–100) across skills before installation. Syncs deterministically via lockfiles and supports private Git repos, npm registries, and 45+ AI agents through platform-specific config files. Built around a modular skill composition system allowing merging, chaining, and conditional loading with integrated testing and sandbox preview capabilities.
About speckit-agent-skills
dceoy/speckit-agent-skills
Agent skills for Spec Kit
Provides reusable skills implementing Spec-Driven Development across multiple agent runtimes (Claude Code, GitHub Copilot CLI, Codex, Gemini) through a shared skill directory with symlinked runtime-specific access points. Skills are declaratively configured via YAML front matter in `SKILL.md` files and orchestrate a seven-stage workflow from constitution through implementation, with optional clarification and validation stages. Integrates with Spec Kit as the primary framework, offering helper scripts and templates in `.specify/` for generating specs, plans, tasks, and checklists from agent interactions.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work