Hermes Agent v0.21.0 'Pantheon' 发布：内置Bot Mode、Agent间直接通信、定时任务记忆

Official release notes (https://github.com/NousResearch/hermes-agent/releases/tag/v2026.8.31) # Hermes Agent v0.21.0 (v2026.8.31) **Release Date:** August 31, 2026 **Since v0.20.0:** ~5,800 commits · ~2,475 merged PRs · ~5,680 files changed · ~869,000 insertions · ~135,000 deletions · **~2,100 issues closed** · 760+ contributors > **The Pantheon Release.** v0.20.0 made Hermes the herald — he spoke, and he carried word to other agents. In v0.21.0 the gods assemble. Bot Mode ships built into the desktop app: a society of named agents with their own faces and group chats, where your bots talk to each other — and to you — like a team, not a toolbox. Around that spine: cron jobs gained memory and continuity so scheduled agents actually learn between runs, subagents can be steered live mid-flight, the MCP surface became a real command center, and the agent can now drive the desktop's own browser. This release rolls up everything from the v0.20.1–v0.20.6 infrastructure patch tags — those windows are fully documented here. --- ## ✨ Highlights - **Bot Mode — your agents become a society, built in** — Bot Mode is now a bundled, default-on part of the desktop app: every agent profile gets a name, a deterministic avatar face (with randomize/lock controls), and a place in a shared roster. Create Discord-style group chats where multiple bots and you talk in one room, @-mention any bot from the composer, and give rooms names and pictures. Before, "multi-agent" meant plumbing; now it looks like a chat app full of coworkers. ([#87886](https://github.com/NousResearch/hermes-agent/pull/87886), [#88243](https://github.com/NousResearch/hermes-agent/pull/88243), [#89386](https://github.com/NousResearch/hermes-agent/pull/89386), [#96726](https://github.com/NousResearch/hermes-agent/pull/96726) — @teknium1, @OutThisLife, @dokterdok) - **`hermes peer` — bot-to-bot DMs between your agents** — Any Hermes agent can now message any other by handle, across profiles and gateways, from the CLI or from inside a conversation. Ask your research bot to hand findings to your coding bot and get the reply back where you can read it. Replies land in each agent's canonical Bot Chat, so conversations between agents are durable and inspectable, not fire-and-forget. ([#88725](https://github.com/NousResearch/hermes-agent/pull/88725), [#88178](https://github.com/NousResearch/hermes-agent/pull/88178), [#91487](https://github.com/NousResearch/hermes-agent/pull/91487) — @teknium1) - **Cron jobs that remember** — Scheduled jobs stopped being goldfish. Cron agents now load and update persistent memory like every other agent, `continuity=true` carries each run's output into the next (so a monitor can dedupe against what it already reported), every job gets a durable notepad scratchpad, monitor-mode jobs skip the LLM entirely when nothing changed, and cron output can land in a bot's canonical Bot Chat — where the bot actually responds. Your 9am briefing job now knows what it told you yesterday. ([#91447](https://github.com/NousResearch/hermes-agent/pull/91447), [#80774](https://github.com/NousResearch/hermes-agent/pull/80774), [#81139](https://github.com/NousResearch/hermes-agent/pull/81139), [#81138](https://github.com/NousResearch/hermes-agent/pull/81138), [#91487](https://github.com/NousResearch/hermes-agent/pull/91487) — @teknium1, @smwbev) - **Steer your subagents while they run** — `delegate_task` gained live orchestration: list running children, steer one mid-flight with a course correction, or stop it early and keep the partial result. Add optional JSON-schema validation on child outputs, per-delegation cost surfaced in results, and raised defaults (250 iterations, 10 concurrent children) — delegation went from fire-and-pray to actually managed parallel work. ([#85232](https://github.com/NousResearch/hermes-agent/pull/85232), [#81144](https://github.com/NousResearch/hermes-agent/pull/81144), [#81142](https://github.com/NousResearch/hermes-agent/pull/81142), [#86506](https://github.com/NousResearch/hermes-agent/pull/86506), [#86745](https://github.com/NousResearch/hermes-agent/pull/86745) — @teknium1) - **The MCP command center** — MCP servers and the catalog merged into one coherent desktop page with drag-in "paste anything" import, background health checks that nudge you to re-auth before a tool call fails, a fleet cost/usage overlay showing schema token estimates and 30-day usage per server, and `hermes://` deep links that install an MCP server with explicit confirmation. Managing twenty MCP servers used to be config-file archaeology; now it's a dashboard. ([#87525](https://github.com/NousResearch/hermes-agent/pull/87525), [#87572](https://github.com/NousResearch/hermes-agent/pull/87572), [#87576](https://github.com/NousResearch/hermes-agent/pull/87576), [#87579](https://github.com/NousResearch/hermes-agent/pull/87579), [#87581](https://github.com/NousResearch/hermes-agent/pull/87581) — @teknium1) - **A CLI power wave** — Ctrl+P opens a fuzzy command palette, the `/model` picker filters as you type, `/status` shows reasoning mode, pending approvals, and context usage, and the status bar can display live cache-hit %, latency, and tokens/sec with per-field toggles. Plus: a global emergency stop, session pin/unpin, rotating task-oriented composer placeholders — and Ghostty-level **terminal pets**, because a companion should have a companion. ([#90730](https://github.com/NousResearch/hermes-agent/pull/90730), [#90717](https://github.com/NousResearch/hermes-agent/pull/90717), [#90745](https://github.com/NousResearch/hermes-agent/pull/90745), [#98250](https://github.com/NousResearch/hermes-agent/pull/98250), [#98282](https://github.com/NousResearch/hermes-agent/pull/98282), [#97666](https://github.com/NousResearch/hermes-agent/pull/97666) — @teknium1, and salvaged community work) - **The agent drives the desktop's browser** — The in-app browser stopped being a window the agent could only look at: Hermes now navigates, clicks, and reads it directly, and pages can be popped out to your system browser with full link context menus. Ask it to walk a docs site or debug a web app and watch it happen inside your own app. ([#90197](https://github.com/NousResearch/hermes-agent/pull/90197), [#89366](https://github.com/NousResearch/hermes-agent/pull/89366) — @OutThisLife, @ethernet8023) - **Six new providers and a model catalog wave** — Meta Model API (Muse Spark) arrives as a built-in provider, alongside CommandCode, Tencent TokenPlan, Nebius Token Factory, Ramp Router, and Actual Computer. The catalogs picked up GLM-5.3-Flash, qwen3.8-max/flash, Gemini 3.7 Flash, MiniMax M3 free, and Nemotron 3.5 Lightning — and `model_overrides` lets you patch any model's context window or pricing yourself without waiting on a release. ([#88565](https://github.com/NousResearch/hermes-agent/pull/88565), [#88308](https://github.com/NousResearch/hermes-agent/pull/88308), [#97917](https://github.com/NousResearch/hermes-agent/pull/97917), [#97916](https://github.com/NousResearch/hermes-agent/pull/97916), [#97915](https://github.com/NousResearch/hermes-agent/pull/97915), [#85560](https://github.com/NousResearch/hermes-agent/pull/85560) — @teknium1 and community) - **Security hardening across the board** — Protected agent-instruction files (AGENTS.md, skills, memory stores) now always require write approval so a prompt-injected agent can't quietly rewrite its own standing orders. A deep redaction sweep closed secret-leak gaps across terminal errors, `.env` file reads, checkpoints, and ACP logs; the approval system learned Windows destructive commands; and macOS permission grants finally survive updates via a stable TCC signing identity. ([#81152](https://github.com/NousResearch/hermes-agent/pull/81152), [#80965](https://github.com/NousResearch/hermes-agent/pull/80965), [#84428](https://github.com/NousResearch/hermes-agent/pull/84428), [#95091](https://github.com/NousResearch/hermes-agent/pull/95091) — @teknium1 and community) --- ## 🏗️ Core Agent & Architecture ### Providers & Models - **New providers:** Actual Computer inference ([#79644](https://github.com/NousResearch/hermes-agent/pull/79644), salvage of #26491), CommandCode with GOAT/Pro/Max plans ([#88308](https://github.com/NousResearch/hermes-agent/pull/88308), salvage of #32909), Meta Model API (Muse Spark) as a built-in provider plugin ([#88565](https://github.com/NousResearch/hermes-agent/pull/88565)), Tencent TokenPlan ([#97917](https://github.com/NousResearch/hermes-agent/pull/97917)), Nebius Token Factory ([#97916](https://github.com/NousResearch/hermes-agent/pull/97916)), and Ramp Router ([#97915](https://github.com/NousResearch/hermes-agent/pull/97915)). - **Model catalog wave:** qwen3.8-max and qwen3.8-flash ([#78024](https://github.com/NousResearch/hermes-agent/pull/78024), [#96979](https://github.com/NousResearch/hermes-agent/pull/96979)), Gemini 3.7 Flash ([#85526](https://github.com/NousResearch/hermes-agent/pull/85526)), GLM-5.3-Flash across OpenRouter/Nous/z.ai/OpenCode ([#95621](https://github.com/NousResearch/hermes-agent/pull/95621), [#96293](https://github.com/NousResearch/hermes-agent/pull/96293)), MiniMax M3 + Inkling free SKUs ([#96264](https://github.com/NousResearch/hermes-agent/pull/96264)), Nemotron 3.5 Lightning ([#84645](https://github.com/NousResearch/hermes-agent/pull/84645)), Meta Muse Spark 1.2 ([#88600](https://github.com/NousResearch/hermes-agent/pull/88600)). - **Per-model metadata overrides** via `model_overrides` config — patch context windows, pricing, or capabilities for any model without waiting on a release ([#85560](https://github.com/NousResearch/hermes-agent/pull/85560)). - **Data-training-tier warnings** — a unified selection-guard registry warns you across every picker surface when a model trains on your data ([#85917](https://github.com/NousResearch/hermes-agent/pull/85917)). - **Pip-installed model providers** discovered via entry points — third parties can ship providers as packages ([#85504](https://github.com/NousResearch/hermes-agent/pull/85504), salvage #81419). - **Prompt caching engaged for LiteLLM Claude** on the OpenAI wire ([#87517](https://github.com/NousResearch/hermes-agent/pull/87517) — @kshitijk4poor) and api.meta.ai routed through the Responses API for caching ([#88601](https://github.com/NousResearch/hermes-agent/pull/88601)). - **Codex GPT context defaults** back to the verified 272K, with explicit -900k picker variants for the large window ([#92797](https://github.com/NousResearch/hermes-agent/pull/92797)). ### Agent Loop & Reliability - **Stalled-provider recovery** — a stalled stream no longer leaves Desktop silent or burns the retry budget ([#87236](https://github.com/NousResearch/hermes-agent/pull/87236) — @fangliquanflq). - **MoA hardening cluster** — the Mixture-of-Agents facade now survives client rebuilds, stream retries, provider rotation, and fallback/restore without leaking prepared requests to native clients ([#90212](https://github.com/NousResearch/hermes-agent/pull/90212) and siblings). - **Structured-output resilience** — auxiliary calls translate `response_format` for Anthropic wires and retry when a provider rejects structured output ([#89589](https://github.com/NousResearch/hermes-agent/pull/89589) — @ethernet8023). - **Compression: lean tail mode** + a compaction recall eval harness to measure what compression actually preserves ([#87326](https://github.com/NousResearch/hermes-agent/pull/87326)). - **Oversized tool results spill to cache** instead of being truncated in sandbox-less sessions ([#89028](https://github.com/NousResearch/hermes-agent/pull/89028)). - **`clarify` asks multiple independent questions in one call** — one form instead of a chain of round-trips ([#89467](https://github.com/NousResearch/hermes-agent/pull/89467) — @ethernet8023). - **Verify subsystem** (port from superagent-ai/grok-cli): run-recipe detection + environment manifest so "it works" claims are backed by detected build/test commands ([#80686](https://github.com/NousResearch/hermes-agent/pull/80686)). ### Sessions & State - **Resumed transcripts never re-append** — `_db_persisted` stamped at row load time ([#92539](https://github.com/NousResearch/hermes-agent/pull/92539) — @kshitijk4poor). - **SessionDB context-manager protocol** for leak-free DB lifecycles ([#88048](https://github.com/NousResearch/hermes-agent/pull/88048) — @jackulau). - **Instant session naming** from the opening message, with sticky titles across resume ([#81985](https://github.com/NousResearch/hermes-agent/pull/81985) — @OutThisLife). - **Session pin/unpin from the CLI** — durable keep flags ([#80761](https://github.com/NousResearch/hermes-agent/pull/80761)). ## 📱 Messaging Platforms (Gateway) - **Slack native live cards** — real streaming replies via chat.startStream plus opt-in plan/task cards for tool progress ([#85476](https://github.com/NousResearch/hermes-agent/pull/85476)); outbound link-preview suppression across native and relay planes ([#95142](https://github.com/NousResearch/hermes-agent/pull/95142) — @benbarclay). - **Telegram inline picker** — every command and skill searchable via @botname, bypassing Telegram's command-menu cap ([#98317](https://github.com/NousResearch/hermes-agent/pull/98317)). - **Relay matured** — native static/dynamic plugin initialization ([#77915](https://github.com/NousResearch/hermes-agent/pull/77915) — @bbednarski9), live-card ops with draft streaming + task cards ([#85796](https://github.com/NousResearch/hermes-agent/pull/85796) — @victor-kyriazakos), session-span segmentation ([#85467](https://github.com/NousResearch/hermes-agent/pull/85467)), voice-note STT restored over the relay lane ([#95274](https://github.com/NousResearch/hermes-agent/pull/95274) — @benbarclay), transport dedupe/fail-fast fixes ([#89584](https://github.com/NousResearch/hermes-agent/pull/89584)). - **Gateway control socket** — fleet consumers query the gateway itself (identify/status), and updaters pause gateways gracefully instead of tree-killing them ([#92447](https://github.com/NousResearch/hermes-agent/pull/92447), [#95695](https://github.com/NousResearch/hermes-agent/pull/95695)). - **Selective multiplex profile serving** ([#83400](https://github.com/NousResearch/hermes-agent/pull/83400)), per-profile MCP lifecycle RPCs ([#86473](https://github.com/NousResearch/hermes-agent/pull/86473)), roster previews showing the latest message ([#85905](https://github.com/NousResearch/hermes-agent/pull/85905)), concise background-process notifications by default ([#85877](https://github.com/NousResearch/hermes-agent/pull/85877)). - **Turn-reaper stack dumps** — when the watchdog fires, wedged worker stacks are captured for diagnosis ([#86248](https://github.com/NousResearch/hermes-agent/pull/86248)), with loop-watchdog tuning knobs wired end-to-end ([#92317](https://github.com/NousResearch/hermes-agent/pull/92317)). - **Split-delivery bug class closed** — swallowed finals and split deliveries fixed as a class ([#79669](https://github.com/NousResearch/hermes-agent/pull/79669) — @kshitijk4poor, salvages #78558). ## 🖥️ Desktop App - **Bot Mode built in** — see Highlights. Full sub-catalog: attributed agent-to-agent message cards ([#85855](https://github.com/NousResearch/hermes-agent/pull/85855)), sender-side delivery notices ([#85888](https://github.com/NousResearch/hermes-agent/pull/85888)), paint-first hydration for instant wakes ([#89510](https://github.com/NousResearch/hermes-agent/pull/89510)), editable group names and room pictures ([#89371](https://github.com/NousResearch/hermes-agent/pull/89371)), Routines pane ([#88731](https://github.com/NousResearch/hermes-agent/pull/88731)), rebuilt on the app design system ([#96726](https://github.com/NousResearch/hermes-agent/pull/96726) — @OutThisLife). - **In-app browser the agent can drive** — the agent uses the desktop's browser tab, not just looks at it ([#90197](https://github.com/NousResearch/hermes-agent/pull/90197) — @OutThisLife), with external-open + link context menus ([#89366](https://github.com/NousResearch/hermes-agent/pull/89366)). - **Detached update hand-off on every OS** — quit → update → reopen with a single shim window; Windows desktop updates no longer park on "Updating Hermes" ([#83634](https://github.com/NousResearch/hermes-agent/pull/83634), [#90937](https://github.com/NousResearch/hermes-agent/pull/90937), [#95897](https://github.com/NousResearch/hermes-agent/pull/95897)). - **Quality-of-life wave** — HUD snaps to cursor on ⌘⇧G ([#83130](https://github.com/NousResearch/hermes-agent/pull/83130)), paste into the composer without focusing it ([#84955](https://github.com/NousResearch/hermes-agent/pull/84955)), per-turn duration badges ([#87245](https://github.com/NousResearch/hermes-agent/pull/87245)), readable macOS window translucency ([#88744](https://github.com/NousResearch/hermes-agent/pull/88744)), Linux launcher entry ([#76456](https://github.com/NousResearch/hermes-agent/pull/76456)), Linux keychain auto-detection ([#84903](https://github.com/NousResearch/hermes-agent/pull/84903)), cron + blueprint recipes in the sidebar ([#85162](https://github.com/NousResearch/hermes-agent/pull/85162)). - **Send Diagnostics** — error cards can upload a redacted debug bundle with a support handoff ([#92020](https://github.com/NousResearch/hermes-agent/pull/92020)). - **Seq-stamped event replay** — lossless desktop reconnect over WebSocket ([#94219](https://github.com/NousResearch/hermes-agent/pull/94219) — @kshitijk4poor). - **God-file decomposition** — the desktop codebase refactored into atomic modules ([#89611](https://github.com/NousResearch/hermes-agent/pull/89611) — @OutThisLife). ## 🖥️ CLI & TUI - See Highlights for the power wave (command palette, fuzzy /model, richer /status, live status-bar metrics, pets). - **`hermes approval-check`** — dry-run approval verdicts for a command without running it ([#81137](https://github.com/NousResearch/hermes-agent/pull/81137)). - **Session titles in status bars** ([#81947](https://github.com/NousResearch/hermes-agent/pull/81947)); subagent model exposed in the auxiliary-models picker ([#80463](https://github.com/NousResearch/hermes-agent/pull/80463)). - **Ctrl+C fixed** — stopped pushing the Kitty keyboard protocol that broke interrupts ([#87074](https://github.com/NousResearch/hermes-agent/pull/87074)). ## 🔧 Tool System & MCP - **MCP command center** — see Highlights ([#87525](https://github.com/NousResearch/hermes-agent/pull/87525) et al.). - **Interrupted commands no longer poison cwd** — the terminal won't adopt a stale working directory after an interrupt ([#85654](https://github.com/NousResearch/hermes-agent/pull/85654) — @ethernet8023). - **Windows rename contention recovery** ([#84852](https://github.com/NousResearch/hermes-agent/pull/84852) — @OutThisLife); colon-bearing session IDs no longer break sandbox mounts ([#93488](https://github.com/NousResearch/hermes-agent/pull/93488), consolidated 6 PRs). - **Vision output uncapped** — aux vision calls stop clamping max_tokens ([#75253](https://github.com/NousResearch/hermes-agent/pull/75253) — @adikpb). - **Subagent work protected** — delegate no longer deletes a child's uncommitted work when git inspection fails ([#88419](https://github.com/NousResearch/hermes-agent/pull/88419) — @kshitijk4poor). ## 🧩 Skills Ecosystem - **Productivity skill wave** (salvaged from @community proposals): document-to-action-items, meeting-action-items, email-inbox-triage, github-issue-to-pr, weekly-review-planning, competitor-news-monitor, product-price-monitor, social-media-content-calendar ([#81185](https://github.com/NousResearch/hermes-agent/pull/81185), [#81749](https://github.com/NousResearch/hermes-agent/pull/81749), [#81672](https://github.com/NousResearch/hermes-agent/pull/81672), [#81708](https://github.com/NousResearch/hermes-agent/pull/81708), [#81936](https://github.com/NousResearch/hermes-agent/pull/81936), [#81945](https://github.com/NousResearch/hermes-agent/pull/81945), [#81906](https://github.com/NousResearch/hermes-agent/pull/81906), [#81942](https://github.com/NousResearch/hermes-agent/pull/81942)). - **HAR-derived API clients** — watch a website once, then call its hidden API directly with no browser ([#70823](https://github.com/NousResearch/hermes-agent/pull/70823)). - **publish-site** — versioned website publishing to GitHub/Cloudflare/Netlify Pages ([#72189](https://github.com/NousResearch/hermes-agent/pull/72189)); **session-librarian** for prompt-driven session management ([#84917](https://github.com/NousResearch/hermes-agent/pull/84917)); **blocked-page-recovery** fallback ladder ([#84906](https://github.com/NousResearch/hermes-agent/pull/84906)); **merge-reconciler** for neutral multi-agent conflict resolution ([#83063](https://github.com/NousResearch/hermes-agent/pull/83063)); Box productivity skill ([#85767](https://github.com/NousResearch/hermes-agent/pull/85767)); ast-grep codemods ([#80892](https://github.com/NousResearch/hermes-agent/pull/80892)); draw-your-font + simple-english optional skills ([#80814](https://github.com/NousResearch/hermes-agent/pull/80814), [#80811](https://github.com/NousResearch/hermes-agent/pull/80811)); plan-interrogation (evolved grill-me) ([#97831](https://github.com/NousResearch/hermes-agent/pull/97831)). - **Advisory SKILL.md linter on create** ([#81896](https://github.com/NousResearch/hermes-agent/pull/81896)); skills-hub live-repo fallback for missing optional skills ([#82780](https://github.com/NousResearch/hermes-agent/pull/82780)). ## ⏰ Cron, Delegation & Multi-Agent - **Cron memory + continuity** — see Highlights ([#91447](https://github.com/NousResearch/hermes-agent/pull/91447), [#80774](https://github.com/NousResearch/hermes-agent/pull/80774)). - **Monitor-mode jobs** with hash-suppressed change detection ([#81138](https://github.com/NousResearch/hermes-agent/pull/81138)); per-job durable notepads ([#81139](https://github.com/NousResearch/hermes-agent/pull/81139)); pre-dispatch config validation ([#81140](https://github.com/NousResearch/hermes-agent/pull/81140)); acked failure signatures stop re-pinging ([#95017](https://github.com/NousResearch/hermes-agent/pull/95017)); per-job reasoning-effort pinning ([#91244](https://github.com/NousResearch/hermes-agent/pull/91244)); "Trigger now" executes immediately and safely ([#70638](https://github.com/NousResearch/hermes-agent/pull/70638) — @smwbev); model-drift impact surfaced in Desktop ([#83266](https://github.com/NousResearch/hermes-agent/pull/83266) — @ctaylor86). - **Live subagent orchestration** — see Highlights ([#85232](https://github.com/NousResearch/hermes-agent/pull/85232)); structured-output schemas on delegate_task ([#81144](https://github.com/NousResearch/hermes-agent/pull/81144)); batch-quality validation before spawning ([#81141](https://github.com/NousResearch/hermes-agent/pull/81141)); truncation markers on capped children ([#86641](https://github.com/NousResearch/hermes-agent/pull/86641)). - **Kanban hardening** — collision-hotspot flagging ([#83428](https://github.com/NousResearch/hermes-agent/pull/83428)), split-brain decision-ownership contract ([#83424](https://github.com/NousResearch/hermes-agent/pull/83424)), memory-aware dispatch guard ([#88115](https://github.com/NousResearch/hermes-agent/pull/88115)), notify/wake delivery modes ([#85487](https://github.com/NousResearch/hermes-agent/pull/85487)). ## 📊 Observability - **Model/provider usage reporting** and bounded tool-metrics aggregation land as an observability subsystem ([#68881](https://github.com/NousResearch/hermes-agent/pull/68881), [#68882](https://github.com/NousResearch/hermes-agent/pull/68882) — @afourniernv), with lifecycle ops bounded so a wedged pipeline can never block the agent ([#83514](https://github.com/NousResearch/hermes-agent/pull/83514) — @victor-kyriazakos). ## 🔒 Security & Reliability - **Protected instruction files** — writes to AGENTS.md/skills/memory always require approval ([#81152](https://github.com/NousResearch/h Repository README (current repository documentation, not release changes) # Hermes Agent ☤ Hermes Agent | Hermes Desktop **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. Use any model you want — [Nous Portal](https://portal.nousresearch.com), OpenRouter, OpenAI, your own endpoint, and [many others](https://hermes-agent.nousresearch.com/docs/integrations/providers). Switch with `hermes model` — no code changes, no lock-in. A real terminal interfaceFull TUI with multiline editing, slash-command autocomplete, conversation history, interrupt-and-redirect, and streaming tool output. Lives where you doTelegram, Discord, Slack, WhatsApp, Signal, and CLI — all from a single gateway process. Voice memo transcription, cross-platform conversation continuity. A 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. Scheduled automationsBuilt-in cron scheduler with delivery to any platform. Daily reports, nightly backups, weekly audits — all in natural language, running unattended. Delegates and parallelizesSpawn isolated subagents for parallel workstreams. Write Python scripts that call tools via RPC, collapsing multi-step pipelines into zero-context-cost turns. Runs anywhere, not just your laptopSeven terminal backends — local, Docker, SSH, Singularity, Modal, Daytona, and Vercel Sandbox. Daytona and Modal offer serverless persistence — your agent's environment hibernates when idle and wakes on demand, costing nearly nothing between sessions. Run it on a $5 VPS or a GPU cluster. Research-readyBatch trajectory generation, trajectory compression for training the next generation of tool-calling models. --- ## Quick Install ### Linux, macOS, WSL2, Termux ```bash curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash ``` ### Windows (native, PowerShell) > **Heads up:** Native Windows runs Hermes without WSL — CLI, gateway, TUI, and tools all work natively. If you'd rather use WSL2, the Linux/macOS one-liner above works there too. Found a bug? Please [file issues](https://github.com/NousResearch/hermes-agent/issues). Run this in PowerShell: ```powershell iex (irm https://hermes-agent.nousresearch.com/install.ps1) ``` The installer handles everything: uv, Python 3.11, Node.js, ripgrep, ffmpeg, **and a portable Git Bash** (MinGit, unpacked to `%LOCALAPPDATA%\hermes\git` — no admin required, completely isolated from any system Git install). Hermes uses this bundled Git Bash to run shell commands. If you already have Git installed, the installer detects it and uses that instead. Otherwise a ~45MB MinGit download is all you need — it won't touch or interfere with any system Git. > **Android / Termux:** The tested manual path is documented in the [Termux guide](https://hermes-agent.nousresearch.com/docs/getting-started/termux). On Termux, Hermes installs a curated `.[termux]` extra because the full `.[all]` extra currently pulls Android-incompatible voice dependencies. > > **Windows:** Native Windows is fully supported — the PowerShell one-liner above installs everything. If you'd rather use WSL2, the Linux command works there too. Native Windows install lives under `%LOCALAPPDATA%\hermes`; WSL2 installs under `~/.hermes` as on Linux. After installation: ```bash source ~/.bashrc # reload shell (or: source ~/.zshrc) hermes # start chatting! ``` ### Troubleshooting #### Windows Defender or antivirus flags `uv.exe` as malware If your antivirus (Bitdefender, Windows Defender, etc.) quarantines `uv.exe` from the Hermes `bin` folder (`%LOCALAPPDATA%\hermes\bin\uv.exe`), this is a **false positive**. The file is Astral's `uv` — the Rust Python package manager Hermes bundles to manage its Python environment. ML-based antivirus engines commonly flag unsigned Rust binaries that download and install packages. **To verify your copy is authentic:** ```powershell # Install GitHub CLI if needed winget install --id GitHub.cli # Login to GitHub gh auth login # Run verification $uv = "$env:LOCALAPPDATA\hermes\bin\uv.exe" $ver = (& $uv --version).Split(' ')[1] [Net.ServicePointManager]::SecurityProtocol = [Net.SecurityProtocolType]::Tls12 $zip = "$env:TEMP\uv.zip" Invoke-WebRequest "https://github.com/astral-sh/uv/releases/download/$ver/uv-x86_64-pc-windows-msvc.zip" -OutFile $zip -UseBasicParsing gh attestation verify $zip --repo astral-sh/uv Expand-Archive $zip "$env:TEMP\uv_x" -Force (Get-FileHash "$env:TEMP\uv_x\uv.exe").Hash -eq (Get-FileHash $uv).Hash ``` If attestation says "Verification succeeded" and the last line prints `True`, you're good. **To whitelist Hermes:** - **Windows Defender:** Run PowerShell as Admin → `Add-MpPreference -ExclusionPath "$env:LOCALAPPDATA\hermes\bin"` - **Bitdefender:** Add an exception in the Bitdefender console (Protection > Antivirus > Settings > Manage Exceptions) - Whitelist the **folder**, not the file hash — Hermes updates `uv` and the hash changes every version For more context, see the upstream Astral reports: [astral-sh/uv#13553](https://github.com/astral-sh/uv/issues/13553), [astral-sh/uv#15011](https://github.com/astral-sh/uv/issues/15011), [astral-sh/uv#10079](https://github.com/astral-sh/uv/issues/10079). --- ## Getting Started ```bash hermes # Interactive CLI — start a conversation hermes model # Choose your LLM provider and model hermes tools # Configure which tools are enabled hermes config set # Set individual config values hermes config get # Print individual config values hermes gateway # Start the messaging gateway (Telegram, Discord, etc.) hermes setup # Run the full setup wizard (configures everything at once) hermes claw migrate # Migrate from OpenClaw (if coming from OpenClaw) hermes update # Update to the latest version hermes doctor # Diagnose any issues ``` 📖 **[Full documentation →](https://hermes-agent.nousresearch.com/docs/)** --- ## Skip the API-key collection — Nous Portal Hermes works with whatever provider you want — that's not changing. But if you'd rather not collect five separate API keys for the model, web search, image generation, TTS, and a cloud browser, **[Nous Portal](https://portal.nousresearch.com)** covers all of them under one subscription: - **300+ models** — pick any of them with `/model ` - **Tool Gateway** — web search (Firecrawl), image generation (FAL), text-to-speech (OpenAI), cloud browser (Browser Use), all routed through your sub. No extra accounts. One command from a fresh install: ```bash hermes setup --portal ``` That logs you in via OAuth, sets Nous as your provider, and turns on the Tool Gateway. Check what's wired up any time with `hermes portal info`. Full details on the [Tool Gateway docs page](https://hermes-agent.nousresearch.com/docs/user-guide/features/tool-gateway). You can still bring your own keys per-tool whenever you want — the gateway is per-backend, not all-or-nothing. --- ## CLI vs Messaging Quick Reference Hermes has two entry points: start the terminal UI with `hermes`, or run the gateway and talk to it from Telegram, Discord, Slack, WhatsApp, Signal, or Email. Once you're in a conversation, many slash commands are shared across both interfaces. | Action | CLI | Messaging platforms | | ------------------------------ | --------------------------------------------- | -------------------------------------------------------------------------------- | | Start chatting | `hermes` | Run `hermes gateway setup` + `hermes gateway start`, then send the bot a message | | Start fresh conversation | `/new` or `/reset` | `/new` or `/reset` | | Change model | `/model [provider:model]` | `/model [provider:model]` | | Set a personality | `/personality [name]` | `/personality [name]` | | Retry or undo the last turn | `/retry`, `/undo` | `/retry`, `/undo` | | Compress context / check usage | `/compress`, `/usage`, `/insights [--days N]` | `/compress`, `/usage`, `/insights [days]` | | Browse skills | `/skills` or `/` | `/` | | Interrupt current work | `Ctrl+C` or send a new message | `/stop` or send a new message | | Platform-specific status | `/platforms` | `/status`, `/sethome` | For the full command lists, see the [CLI guide](https://hermes-agent.nousresearch.com/docs/user-guide/cli) and the [Messaging Gateway guide](https://hermes-agent.nousresearch.com/docs/user-guide/messaging). --- ## Documentation All documentation lives at **[hermes-agent.nousresearch.com/docs](https://hermes-agent.nousresearch.com/docs/)**: | Section | What's Covered | | --------------------------------------------------------------------------------------------------- | ---------------------------------------------------------- | | [Quickstart](https://hermes-agent.nousresearch.com/docs/getting-started/quickstart) | Install → setup → first conversation in 2 minutes | | [CLI Usage](https://hermes-agent.nousresearch.com/docs/user-guide/cli) | Commands, keybindings, personalities, sessions | | [Configuration](https://hermes-agent.nousresearch.com/docs/user-guide/configuration) | Config file, providers, models, all options | | [Messaging Gateway](https://hermes-agent.nousresearch.com/docs/user-guide/messaging) | Telegram, Discord, Slack, WhatsApp, Signal, Home Assistant | | [Security](https://hermes-agent.nousresearch.com/docs/user-guide/security) | Command approval, DM pairing, container isolation | | [Tools & Toolsets](https://hermes-agent.nousresearch.com/docs/user-guide/features/tools) | 40+ tools, toolset system, terminal backends | | [Skills System](https://hermes-agent.nousresearch.com/docs/user-guide/features/skills) | Procedural memory, Skills Hub, creating skills | | [Memory](https://hermes-agent.nousresearch.com/docs/user-guide/features/memory) | Persistent memory, user profiles, best practices | | [MCP Integration](https://hermes-agent.nousresearch.com/docs/user-guide/features/mcp) | Connect any MCP server for extended capabilities | | [Cron Scheduling](https://hermes-agent.nousresearch.com/docs/user-guide/features/cron) | Scheduled tasks with platform delivery | | [Context Files](https://hermes-agent.nousresearch.com/docs/user-guide/features/context-files) | Project context that shapes every conversation | | [Architecture](https://hermes-agent.nousresearch.com/docs/developer-guide/architecture) | Project structure, agent loop, key classes | | [Contributing](https://hermes-agent.nousresearch.com/docs/developer-guide/contributing) | Development setup, PR process, code style | | [CLI Reference](https://hermes-agent.nousresearch.com/docs/reference/cli-commands) | All commands and flags | | [Environment Variables](https://hermes-agent.nousresearch.com/docs/reference/environment-variables) | Complete env var reference | --- ## Migrating from OpenClaw If you're coming from OpenClaw, Hermes can automatically import your settings, memories, skills, and API keys. **During first-time setup:** The setup wizard (`hermes setup`) automatically detects `~/.openclaw` and offers to migrate before configuration begins. **Anytime after install:** ```bash hermes claw migrate # Interactive migration (full preset) hermes claw migrate --dry-run # Preview what would be migrated hermes claw migrate --preset user-data # Migrate without secrets hermes claw migrate --overwrite # Overwrite existing conflicts ``` What gets imported: - **SOUL.md** — persona file - **Memories** — MEMORY.md and USER.md entries - **Skills** — user-created skills → `~/.hermes/skills/openclaw-imports/` - **Command allowlist** — approval patterns - **Messaging settings** — platform configs, allowed users, working directory - **API keys** — allowlisted secrets (Telegram, OpenRouter, OpenAI, Anthropic, ElevenLabs) - **TTS assets** — workspace audio files - **Workspace instructions** — AGENTS.md (with `--workspace-target`) See `hermes claw migrate --help` for all options, or use the `openclaw-migration` skill for an interactive agent-guided migration with dry-run previews. --- ## Contributing We welcome contributions! See the [Contributing Guide](https://hermes-agent.nousresearch.com/docs/developer-guide/contributing) for development setup, code style, and PR process. Quick start for contributors — use the standard installer, then work from the full git checkout it creates at `$HERMES_HOME/hermes-agent` (usually `~/.hermes/hermes-agent`). This matches the layout used by `hermes update`, the managed venv, lazy dependencies, gateway, and docs tooling. ```bash curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash cd "${HERMES_HOME:-$HOME/.hermes}/hermes-agent" uv pip install -e ".[all,dev]" scripts/run_tests.sh ``` Manual clone fallback (for throwaway clones/CI where you intentionally do not want the managed install layout): Create the venv outside the cloned source tree — a venv inside the directory the agent operates from can be wiped by a relative-path command the agent runs against its own checkout, destroying the running runtime mid-session. ```bash curl -LsSf https://astral.sh/uv/install.sh | sh uv venv ~/.hermes/venvs/hermes-dev --python 3.11 source ~/.hermes/venvs/hermes-dev/bin/activate uv pip install -e ".[all,dev]" scripts/run_tests.sh ``` --- ## Community - 💬 [Discord](https://discord.gg/NousResearch) - 📚 [Skills Hub](https://agentskills.io) - 🐛 [Issues](https://github.com/NousResearch/hermes-agent/issues) - 🔌 [computer-use-linux](https://github.com/avifenesh/computer-use-linux) — Linux desktop-control MCP server for Hermes and other MCP hosts, with AT-SPI accessibility trees, Wayland/X11 input, screenshots, and compositor window targeting. - 🔌 [HermesClaw](https://github.com/AaronWong1999/hermesclaw) — Community WeChat bridge: Run Hermes Agent and OpenClaw on the same WeChat account. --- ## License MIT — see [LICENSE](LICENSE). Built by [Nous Research](https://nousresearch.com).