项目简介
Hermes Agent 是由 Nous Research 构建的自我改进 AI 代理。它拥有内置学习循环，能从经验中创建技能并主动持久化知识。支持多种消息平台、任何 LLM 提供商，并可在各种后端上运行。
README 预览
\n \n\n\n# Hermes Agent ☤\n\n\n \n \n \n \n \n\n\n**The self-improving AI agent built by [Nous Research](https://nousresearch.com).** It's the only agent with a built-in learning loop — it creates skills from experience, improves them during use, nudges itself to persist knowledge, searches its own past conversations, and builds a deepening model of who you are across sessions. Run it on a $5 VPS, a GPU cluster, or serverless infrastructure that costs nearly nothing when idle. It's not tied to your laptop — talk to it from Telegram while it works on a cloud VM.\n\nUse any model you want — [Nous Portal](https://portal.nousresearch.com), [OpenRouter](https://openrouter.ai) (200+ models), [NVIDIA NIM](https://build.nvidia.com) (Nemotron), [Xiaomi MiMo](https://platform.xiaomimimo.com), [z.ai/GLM](https://z.ai), [Kimi/Moonshot](https://platform.moonshot.ai), [MiniMax](https://www.minimax.io), [Hugging Face](https://huggingface.co), OpenAI, or your own endpoint. Switch with `hermes model` — no code changes, no lock-in.\n\n\nA real terminal interfaceFull TUI with multiline editing, slash-command autocomplete, conversation history, interrupt-and-redirect, and streaming tool output.\nLives where you doTelegram, Discord, Slack, WhatsApp, Signal, and CLI — all from a single gateway process. Voice memo transcription, cross-platform conversation continuity.\nA closed learning loopAgent-curated memory with periodic nudges. Autonomous skill creation after complex tasks. Skills self-improve during use. FTS5 session search with LLM summarization for cross-session recall. Honcho dialectic user modeling. Compatible with the agentskills.io open standard.\nScheduled automationsBuilt-in cron scheduler with delivery to any platform. Daily reports, nightly backups, weekly audits — all in natural language, running unattended.\nDelegates and parallelizesSpawn isolated subagents for parallel workstreams. Write Python scripts that call tools via RPC, collapsing multi-step pipelines into
常见问题 (3)
故障排除为什么在 Linux 上运行 'hermes desktop' 会出现 'ENOENT: no such file or directory, rename electron -> Hermes' 的错误？
该错误由每用户Electron下载缓存中损坏的Electron发行版缓存zip文件导致。请删除缓存文件并重新运行构建。
- 对于Linux：rm ~/.cache/electron/electron-v*.zip （或指定版本，例如 rm ~/.cache/electron/electron-v40.9.3-linux-x64.zip）
- 对于macOS：rm ~/Library/Caches/electron/*.zip
- 对于自定义缓存的macOS：若设置了XDG_CACHE_HOME或ELECTRON_CACHE，请遵循对应路径。
清除损坏的zip文件后，重新执行'hermes desktop'。构建将重新下载干净的Electron发行版并成功完成。这是一个已知问题，跟踪编号为#39127（自愈缓存清除）。
故障排除为什么Hermes Desktop应用中的聊天滚动条会向后跳动？
这是 useThreadScrollAnchor 逻辑（thread-virtualizer.tsx）中的一个已知错误。ResizeObserver 的解除条件在中间滚动事件中误触发，导致 sticky-bottom 行为暂时失效。该修复已合并到 PR #38221 中，并将包含在 Hermes Desktop v0.15.2 中。在此之前，作为临时解决方案，请尽量减少 composer 的大小调整（例如，关闭 suggestion chips，避免大型代码块重新渲染），或手动将滚动保持在底部。
故障排除为什么Hermes磁盘清理会删除cron/jobs.json，以及如何阻止它？
当旧的 tracked.json 条目（来自 cron 分类修复之前的 Hermes 版本）将 cron/jobs.json 标记为 "cron-output" 时，会导致 disk-cleanup 在 14 天后将其删除。为防止删除，请手动编辑 tracked.json 文件（位于 ~/.hermes/disk-cleanup/tracked.json），移除所有涉及 cron/jobs.json 或 cron/ 目录的条目，然后重启 Hermes。较新版本（在 commit d473e7c93 之后）不会添加此类条目，但已有的陈旧条目必须手动清除。