No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by av · Agent Tool · ★ 101
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is mi safe to install? View the security audit →
agentic coding in 30 loc. a loop, two tools, and an llm. features works with any OpenAI-compatible API: OpenAI, ollama, lmstudio, litellm, vllm, local models tool gives full system access: git, curl, compilers, file I/O (, , heredocs); optional and for background tasks tool spawns a mi subagent with a prompt — runs in the same directory, same API, streams output in real-time; optional tool pursues a high-level goal by iterating subagents until a bash check command exits 0 — uses a progress file so iterations build on prior work (up to 128 by default) recursive agents: and tools spawn sub-agents natively; any tool can also call as a child process self-extending: agent can write its own tools via the skill tool loads markdown playbooks from and (auto-advertised in system prompt) bundled skills: , , , , , , , , , , modular tools: add new tools by dropping files in (hot-loaded before each model call) mode: run inside a Docker container with PWD mounted — safe for destructive experiments. image ships with node, python, uv, git, GNU coreutils/grep/sed/awk/find, curl, jq, ssh, rsync,
| Stars | 101 |
| Forks | 9 |
| Language | JavaScript |
| Category | Agent Tool |
| Quality Score | 70.5640904450584/100 |
| Open Issues | 1 |
| Last Updated | 2026-07-02 |
| Created | 2026-04-13 |
| Platforms | cli, node |
| Est. Tokens | ~16k |
These tools work well together with mi for enhanced workflows:
Looking for a mi alternative? If you're comparing mi with other agent tool tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Any Agent × Any Provider × Anywhere. Connect Cursor, Claude Code, OpenCode, or Codex to Slack, Telegram, Disco
The easiest way to run multiple Claude Code sessions, each in its own container, with a dashboard to manage th
Open-source Agentic AI framework in Go for building, orchestrating, and deploying intelligent agents. LLM-agno
Open source AI coding platform with Web IDE, multi-agent system, 37+ tools, MCP protocol. MIT licensed.
A terminal coding agent, and a Python SDK for embedding on-device models in your own apps.
Run Claude agents in secure cloud sandboxes — via API, CLI, or Slack. One call. Full agent. Zero infrastructur
Explore other popular agent tool tools:
mi is self-extending sandboxed agentic harness. a loop, two tools and an llm. It is categorized as a Agent Tool with 101 GitHub stars.
mi is primarily written in JavaScript. It covers topics such as agentic-ai, agentic-coding, ai.
You can find installation instructions and usage details in the mi GitHub repository at github.com/av/mi. The project has 101 stars and 9 forks, indicating an active community.
The top alternatives to mi on Agent Skills Hub include golembot, safeclaw, AgenticGoKit. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: