AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Workflow Automation/
hermes-agent
hermes-agent logo

hermes-agent

Active·★ 256.0k·MIT·Updated 2026-10-22
★ Most Popular

Hermes Agent is Nous Research's general-purpose agent, built around what the README calls a closed learning loop. After a complex task it can create a skill, refine that skill as it is reused, nudge itself to save knowledge, search its own past conversations with SQLite FTS5 plus LLM summaries, and build a user model through Honcho. Skills follow the agentskills.io open standard, so they are plain files you can read and edit.

It is designed to live on a server rather than your laptop. One gateway process serves a terminal UI and Telegram, Discord, Slack, WhatsApp and Signal, with voice-memo transcription and conversation continuity across them. A built-in cron scheduler runs natural-language jobs and delivers results to any of those channels, and seven terminal backends (local, Docker, SSH, Singularity, Modal, Daytona, Vercel Sandbox) decide where its commands actually execute.

It is not tied to one model: Nous Portal, OpenRouter, OpenAI or your own endpoint can be switched with hermes model. It can also spawn isolated subagents and generate tool-calling trajectories for training research. Install is a one-line script on Linux, macOS and WSL2, with a native PowerShell installer for Windows. License is MIT.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool.

#agent-framework#ai#ai-agent#ai-agents#anthropic#chatgpt#claude#claude-code
$ Install
$ curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
↗ Visit site★ GitHub
01

Features

01Autonomous skill creation after complex tasks, with skills that self-improve during use
02FTS5 session search with LLM summarization and Honcho user modeling for cross-session recall
03Single gateway for Telegram, Discord, Slack, WhatsApp, Signal and CLI
04Built-in cron scheduler with delivery to any platform
05Seven terminal backends: local, Docker, SSH, Singularity, Modal, Daytona, Vercel Sandbox
06Switch model providers with hermes model, no code changes
02

Why choose it

+Learns from use: it turns completed tasks into skills and keeps refining them, so repeated work gets faster without manual prompt files.
+One gateway covers terminal, Telegram, Discord, Slack, WhatsApp and Signal, so you can hand it work from your phone while it runs on a server.
+Model-agnostic and MIT-licensed, with serverless backends (Modal, Daytona) that hibernate when idle to keep hosting costs low.
03

Trade-offs

!Self-created skills and persistent memory mean behaviour drifts over time; you need to review what it saves.
!Giving an always-on agent shell backends plus messaging access is a large attack surface that needs careful sandboxing.
!The feature set is broad (gateway, cron, subagents, seven backends), so setup and troubleshooting take more effort than a single CLI agent.
04

Maintenance check

ActiveChecked: 2026-10-22
Last commit
2026-10-22
Commits, last 90 days
100+
Releases, last 90 days
21
Latest release
v0.22.4 (2026-10-21)
Open issues
50391
Recent issues with at least one comment
10/16
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #78647 Repo-wide godfile eradication: residual 2K tasks after #10211782 comments · 2026-08-04
  • #127665 [Bug]: Desktop renders one reply twice while its row is already committed (the overlay's fold exempts pending live rows)58 comments · 2026-09-29
  • #131859 Cannot open a pull request via API: CreatePullRequest permission error (issue creation and fork PRs still work)47 comments · 2026-10-03

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

05

Compatibility

Linux / macOS / WSL2 / Windows
CLI + TUI
Verified via docs
Telegram / Discord / Slack / WhatsApp / Signal
Messaging gateway
Verified via docs
OpenRouter / OpenAI / Nous Portal
Model providers
Verified via docs
06

Quick start

1
$ curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
07

Use cases

↳A personal assistant on a cheap VPS that you talk to from Telegram
↳Unattended daily reports, backups and audits defined in natural language
↳Generating tool-calling trajectories for model training research
08

Alternatives

⚡Top VS:
VSapra-fleet★105→VSTouchpoint★47→VSTrueMemory★381→VSclaude-code-source-all-in-one★91→
apra-fleet logo
apra-fleet★ 105
AI agents that write code, review each other's work, and coordinate across your machines
VSCompare→
Touchpoint logo
Touchpoint★ 47
Give your AI agent eyes and hands on any desktop — cross-platform accessibility API with MCP server
VSCompare→
TrueMemory logo
TrueMemory★ 381
A living memory system that ingests long-horizon data to infer insights, enabling more decisive action, all while running on a single SQLite file locally.
VSCompare→
claude-code-source-all-in-one logo
claude-code-source-all-in-one★ 91
Always up-to-date open-source mirror of Claude Code (currently v2.1.123). Run from source with Claude subscription/API, ChatGPT subscription (GPT-5.5 / GPT-5.4), OpenAI-compatible providers, or local Ollama. 18 deep-dive architecture articles included. For educational and reference purposes only.
VSCompare→
widemem-ai logo
widemem-ai★ 49
Next-gen AI memory layer with importance scoring, temporal decay, hierarchical memory, and YMYL prioritization
VSCompare→
Genexus18MCP logo
Genexus18MCP★ 38
GeneXus 18 MCP server for Claude, Cursor, and AI agents — read, edit, analyze KB objects (transactions, web panels, procedures, SDTs) over the Model Context Protocol.
VSCompare→
cockpit logo
cockpit★ 16
A web UI for Claude Code that frees it from your terminal
VSCompare→
cindy logo
cindy★ 3.2k
Consider it done. The open-source AI agent that works out of the box · 想到,就能做到。开源、开箱即用的 AI Agent。
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Maintenance check05Compatibility06Quick start07Use cases08Alternatives
NousResearch /
hermes-agent
↗

A self-hosted agent that writes its own skills and remembers you across sessions

256.0kMIT
Last commitToday
StatusActive
CategoryWorkflow Automation
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.