
Security News
New Study Identifies 53 Slopsquatting Targets Across 5 Frontier LLMs
Five frontier LLMs generated the same nonexistent package names, leaving 53 available for potential slopsquatting across PyPI and npm.
@open330/agt
Advanced tools
A modular toolkit for extending AI coding agents (Claude Code, Codex CLI, Gemini CLI)
Install and manage skills, personas, and hooks for AI coding agents. Each skill is a self-contained markdown module that gives an agent domain-specific capabilities.
# No install needed — use npx
npx @open330/agt skill install --profile core
npx @open330/agt persona install -g --from jiunbae/agent-skills
# Or install globally
npm install -g @open330/agt
agt skill install --profile core
agt persona install -g --all
agt skill — Skill Managementagt skill install kubernetes-skill # Install to current workspace
agt skill install -g git-commit-pr # Install globally
agt skill install --profile core # Install a profile (curated set)
agt skill install --from user/repo # Install from a remote skills repo
agt skill list # List all available skills
agt skill list --installed # Show installed only
agt skill update # Update all remote-installed skills
agt skill uninstall kubernetes-skill # Remove a skill
agt persona — Persona Managementagt persona list # List personas
agt persona install -g security-reviewer # Install globally
agt persona install -g --all # Install all personas
agt persona create rust-expert --ai "Rust unsafe" # AI-generated persona
agt persona review security-reviewer # Run code review with persona
agt persona review security-reviewer --codex # Use Codex CLI
agt persona show security-reviewer # View persona details
agt run — Skill Executionagt run "scan for security issues" # Auto skill matching
agt run --skill security-auditor "scan" # Specify skill explicitly
agt hook — Hook Managementagt hook install english-coach # Install a hook
agt hook list # List available hooks
agt team — Multi-Agent Teamsagt team run review # Run a team of agents
agt team list # List available teams
Installed by default with --profile core:
| Skill | Description |
|---|---|
git-commit-pr | Git commit & PR guide with security validation |
context-manager | Project context auto-loader |
static-index | Global static context index |
security-auditor | Repository security audit |
background-implementer | Parallel multi-LLM implementation |
background-planner | Parallel multi-LLM planning |
background-reviewer | Multi-LLM parallel code review |
33 skills across 8 categories:
| Category | Skills |
|---|---|
| Agents | background-implementer, background-planner, background-reviewer |
| Development | context-worktree, git-commit-pr, iac-deploy-prep, multi-ai-code-review, playwright, pr-review-loop, task-master |
| Business | bm-analyzer, document-processor, proposal-analyzer |
| Integrations | appstore-connect, discord, google-search-console, kubernetes, notion-summary, obsidian-tasks, obsidian-writer, service-manager, slack, vault-secrets |
| ML/AI | audio-processor, ml-benchmark, model-sync, triton-deploy |
| Security | security-auditor |
| Context | context-manager, static-index |
| Meta | karpathy-guide, skill-manager, skill-recommender |
7 expert personas for AI-powered code review:
| Persona | Domain |
|---|---|
security-reviewer | OWASP, auth, injection |
architecture-reviewer | SOLID, API design, coupling |
code-quality-reviewer | Readability, complexity, DRY |
performance-reviewer | Memory, CPU, I/O, scalability |
database-reviewer | Query optimization, schema, indexing |
frontend-reviewer | React, accessibility, performance |
devops-reviewer | K8s, IaC, CI/CD |
You can create your own skills repository that agt discovers automatically:
my-skills/
├── agents/
│ └── my-agent/
│ └── SKILL.md
├── development/
│ └── my-tool/
│ └── SKILL.md
├── personas/
│ └── my-reviewer/
│ └── PERSONA.md
├── profiles.yml
├── agt.toml
└── README.md
# Others install from your repo
agt skill install --from you/my-skills
agt persona install --from you/my-skills
See full documentation for SKILL.md format, PERSONA.md format, profiles.yml, and agt.toml specifications.
| Platform | Architecture |
|---|---|
| macOS | arm64 (Apple Silicon), x64 (Intel) |
| Linux | x64, arm64 |
Binary is automatically downloaded during npm install. If the optional platform package fails, a fallback download from GitHub Releases is attempted.
MIT
FAQs
agt — A modular toolkit for extending AI coding agents
The npm package @open330/agt receives a total of 7 weekly downloads. As such, @open330/agt popularity was classified as not popular.
We found that @open330/agt demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Security News
Five frontier LLMs generated the same nonexistent package names, leaving 53 available for potential slopsquatting across PyPI and npm.

Security News
The White House’s Gold Eagle Initiative aims to coordinate AI-discovered vulnerabilities, validate findings, and accelerate patching across critical software.

Security News
A Shai-Hulud infection exposed Suno's source code, which shows the AI music startup stream-ripped tracks to train its models.