middleman — security grade SAFE, quality 67/100

Security audit verdict: SAFE · quality 67/100

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 SawyerHood · Agent Tool · ★ 180

Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h

🔒 Is middleman safe to install? View the security audit →

About middleman

[!IMPORTANT] Middleman is being superseded by bb, an agentic IDE that orchestrates Claude Code, Codex, and Pi in long-lived threads you can follow live and steer at any turn. Try it with , then open . Requires Node 22.12.0+ on macOS/Linux, or Windows via WSL2. 👔 Middleman A local-first multi-agent orchestration platform. One manager, many workers, zero tab-juggling. [!WARNING] Middleman is pre-v1, experimental, and updated constantly. Expect breaking changes. If you're using agentic coding tools, you've probably hit this wall: you start with one agent, then two, then five. You're branching, worktree-ing, reviewing, merging, context-switching. The agents are cranking out code. But your entire day is spent managing them, sequencing work, checking output, nudging things along. You're not an IC anymore. You've become a project manager. You need a middle manager. Middleman gives you one. You talk to a single persistent manager agent per project. You describe what needs to be done, whether it's a feature, a batch of bug fixes, or a refactor, and the manager dispatches workers, parallelizes where it makes sense, and surfaces only the things that need your attention.

agentsai-agentsclaude-code

Quick Facts

Stars180
Forks21
LanguageTypeScript
CategoryAgent Tool
LicenseApache-2.0
Quality Score67.2403997223853/100
Open Issues5
Last Updated2026-05-21
Created2026-02-14
Platformsclaude-code, node
Est. Tokens~477k

middleman alternative? Top 6 similar tools

Looking for a middleman alternative? If you're comparing middleman 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.

  • vllora by vllora · ⭐ 812

    Debug your AI agents

  • vestige by samvallad33 · ⭐ 628

    Cognitive Deterministic Memory Security OS for Agentic AI. Deterministic root-cause retrieval that reaches bac

  • claude-code-mastery by TheDecipherist · ⭐ 548

    The complete guide to Claude Code: CLAUDE.md, hooks, skills, MCP servers, and commands

  • kindly-web-search-mcp-server by Shelpuk-AI-Technology-Consulting · ⭐ 388

    Kindly Web Search MCP Server: Web search + robust content retrieval for AI coding tools (Claude Code, Codex, C

  • daymon by daymonio · ⭐ 365

    Daymon puts your favorite AI to work 24/7. It schedules, remembers, and orchestrates your own virtual team. Fr

  • mcp-server-code-execution-mode by elusznik · ⭐ 337

    An MCP server that executes Python code in isolated rootless containers with optional MCP server proxying. Imp

More Agent Tool Tools

Explore other popular agent tool tools:

View all Agent Tool tools →

Popular TypeScript Agent Tools

Frequently Asked Questions

What is middleman?

middleman is The middle manager for your agents. It is categorized as a Agent Tool with 180 GitHub stars.

What programming language is middleman written in?

middleman is primarily written in TypeScript. It covers topics such as agents, ai-agents, claude-code.

How do I install or use middleman?

You can find installation instructions and usage details in the middleman GitHub repository at github.com/SawyerHood/middleman. The project has 180 stars and 21 forks, indicating an active community.

What license does middleman use?

middleman is released under the Apache-2.0 license, making it free to use and modify according to the license terms.

What are the best alternatives to middleman?

The top alternatives to middleman on Agent Skills Hub include vllora, vestige, claude-code-mastery. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.

How this security grade is produced

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:

View on GitHub → Browse Agent Tool tools