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 PromptPartner · Codex Skill · ★ 173
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is agentsmith safe to install? View the security audit →
AgentSmith — the universal agent harness Vibe coding gets you a working demo. AI-assisted engineering is what keeps it standing once customers arrive, the product changes, or another engineer has to touch it. Agentsmith is that second layer — a portable, battle-tested operating system for Claude Code (and any agent that reads a / / ). Drop it on any machine, pick the kind of work you're doing, and the same disciplined core adapts to it — software, devops/setup, marketing & outreach, document creation, data crunching, research, design, and general admin. The agent is the model plus the harness. The model is 10% of the outcome; the harness — the rules, tools, memory, guardrails, and feedback loops around it — is the other 90%, and it's the part you control. This repo is that 90%, generalized and reusable. (The reasoning is in ; the public work these ideas build on is credited in full under principles & influences.) It didn't come from a whiteboard. It grew over 6 months of real, autonomous work — data crunching, marketing outreach, dozens of smaller projects, and one software product still in active development after half a year — across hundreds of sessions.
| Stars | 173 |
| Forks | 6 |
| Language | Python |
| Category | Codex Skill |
| License | MIT |
| Quality Score | 67.8826496301682/100 |
| Open Issues | 2 |
| Last Updated | 2026-09-19 |
| Created | 2026-07-16 |
| Platforms | claude-code, codex, gemini, python |
| Est. Tokens | ~17k |
These tools work well together with agentsmith for enhanced workflows:
Looking for a agentsmith alternative? If you're comparing agentsmith with other codex skill tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Claude Code Skills Marketplace: plugins, skills for ADR-driven development, DevOps automation, ClickHouse mana
A collection of AI agent skills for Clawdbot, Claude Code, Codex
HeyClaude (formerly Claude Pro Directory) is a searchable collection of pre-built AI skills, agents, MCP serve
The Ultimate Claude Code Toolkit — 127 skills, 86 agents, 109 marketplace repos (11,700+ community skills), 30
Overture is an open-source, locally running web interface delivered as an MCP (Model Context Protocol) server
Spec-driven development with smart compaction. Claude Code plugin combining Ralph Wiggum loop with structured
Explore other popular codex skill tools:
agentsmith is Universal, model-agnostic operating harness for AI agents (Claude, Codex, Gemini, …) — a lean core + work-type profiles assembled by one setup script.. It is categorized as a Codex Skill with 173 GitHub stars.
agentsmith is primarily written in Python. It covers topics such as agentic-ai, agentic-coding, ai-agents.
You can find installation instructions and usage details in the agentsmith GitHub repository at github.com/PromptPartner/agentsmith. The project has 173 stars and 6 forks, indicating an active community.
agentsmith is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to agentsmith on Agent Skills Hub include cc-skills, agent-skills, claudepro-directory. 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: