AG Kit
An AI agent toolkit of markdown-based templates for Skills, Agents, and Workflows, with a coordinator mode, persistent memory, and context compression.
Last release Jun 7, 2026 · Last commit Jun 8, 2026 · ★ 7,689
Overview
AG Kit (AGKIT) is an agent toolkit providing templates for Skills, Agents, and Workflows, featuring a coordinator mode for multi-agent orchestration, persistent memory, context compression, and conditional skill loading. It installs a local config into a project. Popularity figures cited in some source material could not be confirmed and are treated as unverified. This is a general dev/agent-tooling project, not onchain-specific.
Details
- Chains
- Unknown
- Agent types
- Agent Framework, Templates
- Open source
- Yes
- Last checked
- 2026-06-12
Marketplace Signals
No marketplace signals recorded for this resource yet.
Verification status reflects evidence reviewed by Sato Hub. Self-reported claims are never presented as verified.
Recent Activity
All news →Releases and posts from AG Kit, pulled from its GitHub and X.
ag-kit 2026.6.7
Documentation site overhaul. Added Multi-language docs (en / vi / 中文 / 日本語) with a flag-dropdown language switcher; all chrome, 7 docs pages, and 10 guide examples translated. Changelog page at /docs/changelog . Working "On This Page" table of contents with scroll-spy. Syntax highlighting (shiki) + web fonts (Inter + JetBrains Mono). Follow-on-X dialog ( @antigravitykit ). Changed Redesigned MDX c
ag-kit AG Kit 2026.6.2 — Full-kit accuracy audit
Full-kit accuracy audit. Reviewed every component of AG Kit — all 45 skills, 20 agents, 13 workflows, the GEMINI.md rules, and both overview docs — against verified 2026 facts (OWASP Top 10:2025, MCP spec 2025-06-18, Rust 2024 edition, Node 24 LTS). Removed dead references, fabricated numbers, contradictions, broken formatting, and stale date stamps. The CLI now surfaces update notifications from
Related Resources
Coinbase AgentKit
ActiveCoinbase's toolkit for giving AI agents wallets and the ability to take onchain actions.
ElizaOS
ActiveOpen-source TypeScript framework for building crypto-native multi-agent systems.
Model Context Protocol (MCP)
ActiveAnthropic's open standard for connecting AI systems to external tools and data sources.
Join the Sato Hub Briefing
One email a week — the agents, tools, and infrastructure that actually shipped, and why they matter.