by phibrowser · Agent Tool · ★ 228
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
None of the 42 rules matched — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against 42 published rules built on the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
The quality score combines nine heuristic checks of the repo and its README: completeness, clarity, specificity, examples, structure, agent readiness, step-by-step content, instruction quality and security notes. It rates how well the project explains itself to an agent, not the code.
| Stars | 228 |
| Forks | 19 |
| Language | Swift |
| Category | Agent Tool |
| License | Apache-2.0 |
| Quality Score | 61/100 |
| Open Issues | 6 |
| Last Updated | 2026-10-10 |
| Created | 2026-03-30 |
| Platforms | browser |
| Est. Tokens | ~17k |
Phi Browser The AI browser for macOS. Agentic, local-first, native Swift. Phi is a Chromium-based browser built as a real native macOS app (AppKit + SwiftUI), with an AI agent that acts in the page, a memory that lives in a file you can read, and the option to run AI entirely on your own machine.
Looking for a phibrowser-mac alternative? If you're comparing phibrowser-mac 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.
Ask several AI models one question — they answer in parallel, critique each other blind, and you decide. Nativ
AI-Human symbiotic browser — browse the web as one entity
Cover your Mac's screen with a hotkey while AI agents keep running, even with the lid closed. The covered scre
Open-source status companion for your AI coding agents — Claude Code, Codex, Antigravity, Grok, and Cursor. Li
macOS 26+ browser CLI for agents: persistent sessions, compact JSON, screenshots, clicks/forms, JS eval, and l
A macOS desktop AI pet that mirrors your Claude Code & Codex CLI session state in real time. 支持Claude Code和Cod
Explore other popular agent tool tools:
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy: 42 published rules, covering credential harvesting, data exfiltration, curl | sh installers and more. 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: