by xianyu-sheng · MCP Server · ★ 56
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
pip install -U xenon-agent
pip install -U "git+https://github.com/xianyu-sheng/Xenon.git@v0.8.5"Taken from the project's README on GitHub.
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.
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 | 56 |
| Forks | 11 |
| Language | Python |
| Category | MCP Server |
| License | MIT |
| Quality Score | 60/100 |
| Open Issues | 21 |
| Last Updated | 2026-11-05 |
| Created | 2026-06-03 |
| Platforms | cli, mcp, python |
| Est. Tokens | ~15k |
Xenon Agent Harness for AI coding agents — 可信、可验证、可评测的 AI Agent 运行时 Xenon 不是又一个 AI 编程助手，而是让 Agent 可信地运行所需的基础设施。 核心差异： 🔒 证据导向架构：工具输出是 Evidence，LLM 输出只是 Claim — 验证闭环抑制幻觉 🛡️ 执行隔离边界：路径围栏、命令注入拦截、权限门 — 所有副作用经同一收敛点 🔄 7 种推理范式：ReAct / Plan-Execute / Reflection 及其组合 — 可替换、可观察 📊 可复现评测：SWE-bench Lite 40.0% 通过率（同模型 +6.7pp），交互与评测共享约束 🎯 生产就绪：v0.8.5 经系统性边界探测，所有逃逸路径均已修复并锁定回归测试 可以当命令行工具直接用，也可以当库嵌入你的 Agent...
Looking for a Xenon alternative? If you're comparing Xenon with other mcp server tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
An agent runtime for Python. Give your agent real work. Keep control: every action checked before it runs, eve
Local-first, auditable Python code agent. Ships its own 30-task hidden-test benchmark plus SWE-bench Verified
Evidence-gated runner for Codex, Claude Code, OpenCode, and local coding agents. Routes tasks into scoped DAG
Turn feature specs into merged PRs with a self-supervising swarm of coding agents — parallel execution, isolat
A memory-first AI agent that remembers why decisions were made — not just the last message. Runs local (Ollama
Outcome-based persistent memory MCP server for Claude Code and OpenCode. Good advice promoted, bad advice demo
Explore other popular mcp server 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: