open-agent-kit
Health Uyari
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 8 GitHub stars
Code Gecti
- Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Gecti
- Permissions — No dangerous permissions requested
This MCP server and CLI tool acts as a team memory for AI-assisted coding. It automatically captures development context, architectural decisions, and plans so that AI agents can retain intelligence between sessions.
Security Assessment
Overall Risk: Medium. While the automated code scan found no dangerous patterns or hardcoded secrets, the underlying architecture inherently requires high system privileges to function. The daemon intercepts inputs via local hooks and executes autonomous agents, which necessitates running continuous background processes. Furthermore, standard installation relies on piping a remote shell script directly to execution (`curl | sh`). Developers must implicitly trust that the remote server and codebase are secure, as any autonomous agent action could potentially interact with the filesystem or network.
Quality Assessment
The project explicitly states it is in maintenance mode and is no longer under active development, having been succeeded by a newer project called Myco. It uses the permissive MIT license and has excellent, detailed documentation. However, community adoption is currently very low, with only 8 GitHub stars. Because it is a memory-capture daemon designed to build an index of your proprietary code, the low visibility means it has not been battle-tested or broadly audited by the open-source community.
Verdict
Use with caution — the tool provides a useful feature set but carries a medium risk due to its low community visibility, its access to your proprietary codebase, and its abandoned maintenance status.
| Maintenance Mode | OAK automatically captures the plans, decisions, and trade-offs that usually vanish. Intelligence powered memories, context injection, autonomous agents, and more.
Open Agent Kit
OAK is in maintenance mode. This project is no longer under active development. I have since started Myco, a new project originally inspired by ideas explored in OAK. Myco goes beyond memory and unlocks true intelligence for your project and your team.
Your Team's Memory in the Age of AI-Written Code
You architect. AI agents build. But the reasoning, trade-offs, and lessons learned disappear between sessions. OAK records the full development story — plans, decisions, gotchas, and context — creating a history that's semantically richer than git could ever be. Then autonomous OAK Agents and Skills turn that captured intelligence into better documentation, deeper insights, and ultimately higher quality software, faster.

graph LR
A[AI Coding Agent] -->|Hooks| B[OAK Daemon]
B -->|Context| A
B --> C[(Memory & Code Index)]
C --> D[OAK Agents]
D -->|Docs · Analysis · Insights| E[Your Project]
Legacy install (maintenance mode)
Quick Start
# Install via Homebrew (macOS)
brew install goondocks-co/oak/oak-ci
# Or via the install script (macOS / Linux)
curl -fsSL https://raw.githubusercontent.com/goondocks-co/open-agent-kit/main/install.sh | sh
# Initialize your project
oak init
Beta / pre-release channelWindows? See QUICKSTART.md for PowerShell install and other methods (pipx, uv, pip).
After installing OAK, switch to the beta channel from the daemon UI
(About dialog) or by editing ~/.oak/update.yaml:
update:
channel: beta
The daemon will automatically check for and download pre-release versions.
Open the OAK Dashboard in your browser:
oak team start --open
Start coding!
claude
Supported Agents
| Agent | Hooks | MCP | Skills |
|---|---|---|---|
| Claude Code | Yes | Yes | Yes |
| Gemini CLI | Yes | Yes | Yes |
| Cursor | Yes | Yes | Yes |
| Codex CLI | Yes (OTel) | Yes | Yes |
| OpenCode | Yes (Plugin) | Yes | Yes |
| Windsurf | Yes | No | Yes |
| VS Code Copilot | Yes | Yes | Yes |
Contributing
See CONTRIBUTING.md for the contributor guide and oak/constitution.md for coding standards.
Security
See SECURITY.md for the vulnerability reporting policy.
License
Yorumlar (0)
Yorum birakmak icin giris yap.
Yorum birakSonuc bulunamadi