Skillget

ai-native-toolkit

agent
Security Audit
Fail
Health Warn
  • License — License: MIT
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 7 GitHub stars
Code Fail
  • rm -rf — Recursive force deletion command in demo/scripts/demo-1-ctx-init.sh
  • rm -rf — Recursive force deletion command in demo/scripts/demo-2-review.sh
  • network request — Outbound network request in demo/vulnerable-saas/components/NotesList.tsx
Permissions Pass
  • Permissions — No dangerous permissions requested

No AI report is available for this listing yet.

SUMMARY

Open-source AI-native engineering plugins for Claude Code — project context, code review, security audit, task clarification, and more

README.md

AI Native Toolkit

Open-source plugins that turn vibe-coding into production-grade AI-native engineering.
Built for Claude Code; most plugins are runtime-agnostic and work in opencode too.

Quick Start

Claude Code

/plugin marketplace add OXI-717/ai-native-toolkit
/plugin install ctx@ai-native-toolkit

opencode

There is no marketplace install for opencode. Clone the repo and point skills.paths
at the plugins you want — skills are picked up recursively:

git clone https://github.com/OXI-717/ai-native-toolkit.git ~/ai-native-toolkit

// ~/.config/opencode/opencode.json
{
  "skills": {
    "paths": [
      "~/ai-native-toolkit/plugins/review",
      "~/ai-native-toolkit/plugins/pentest",
      "~/ai-native-toolkit/plugins/gh-issues",
      "~/ai-native-toolkit/plugins/infocompressor",
      "~/ai-native-toolkit/plugins/deep-interview",
      "~/ai-native-toolkit/plugins/ctx"
    ]
  }
}

Restart opencode afterwards. Plugins not listed above depend on Claude Code hooks
and will not work — see Runtime support.

Plugins

Plugin What it does
ctx Project context: AGENTS.md, rules, init, lint
review Multi-agent code review with confidence filtering
pentest Black-box security audit (L0-L3)
context-handoff Preserve context across /compact and /clear
statusline Claude Code status bar with usage limits
gh-issues GitHub Issues as AI session memory
infocompressor Dense reference specs from long documents
deep-interview Clarify vague tasks before planning
screencast Screencast demo videos of web-app flows
agent-teams Multi-agent team coordination
notebooklm Source-grounded answers from NotebookLM
cc-analytics Claude Code usage analytics
openrouter-setup OpenRouter as an OpenAI-compatible endpoint

Runtime support

Claude Code is the reference runtime: everything works there. opencode loads skills
from skills.paths but does not execute plugin hooks, so anything whose value
comes from lifecycle hooks degrades or stops working.

Plugin Claude Code opencode Note
review full full Skill plus gh CLI, no runtime-specific APIs
pentest full full Recon needs a Playwright MCP server, configured separately
gh-issues full full gh CLI and plain files only
infocompressor full full Pure skill
deep-interview full full Pure skill
ctx full partial ctx-init and ctx-lint work; auto-loading AGENTS.md and rules needs a SessionStart hook
context-handoff full manual Save/load via the skill still works; auto-restore across sessions needs hooks
statusline full no Reads Claude Code status-bar config and usage data; no equivalent source in opencode

ctx in this repo is the public edition: ctx-init and ctx-lint only. The vault,
meetings, people and research skills depend on private infrastructure and are not
exported.

4 Levels of AI-Native Development

Level 1: Vibe Coding — ask ChatGPT, paste code, hope it works

Level 2: Context & Rules → ctx, context-handoff, infocompressor, deep-interview — AI keeps your project and the task itself in focus between sessions

Level 3: Verified Development → review, pentest — AI reviews and audits your code

Level 4: Autonomous Agents → gh-issues, statusline — session state and visibility for long autonomous runs (statusline is Claude Code only)

License

MIT

Reviews (0)

No results found