MCPInstall

awesome-hermes-agent-ecosystem

Navigate the Hermes Agent ecosystem — skills, tools, integrations, deployment, and multi-agent orchestration resources

Installs
176
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/reason-machines/ai-agent-skills --skill awesome-hermes-agent-ecosystem
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads awesome-hermes-agent-ecosystem automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

Awesome Hermes Agent Ecosystem

Skill by ara.so — AI Agent Skills collection.

This skill provides comprehensive knowledge of the Hermes Agent ecosystem maintained at 0xNyk/awesome-hermes-agent. Hermes Agent by Nous Research is a self-improving AI agent with built-in learning loop, autonomous skill curation, and 18-platform messaging support.

What is Hermes Agent?

Hermes Agent is the only agent with:

  • Built-in learning loop — creates skills from experience and improves them during use
  • Autonomous Curator (v0.12.0+) — maintains its own skill library with 7-day grade/consolidate/prune cycles
  • 18 messaging platforms — Telegram, Discord, Slack, WhatsApp, Signal, Feishu/Lark, WeCom, QQBot, Yuanbao, and more
  • 7 terminal backends — local, Docker, SSH, Singularity, Modal, Daytona, Vercel Sandbox
  • Conversational memory — searches past conversations and builds user model across sessions

Core repository: NousResearch/hermes-agent (134k+ stars)

Quick Start Path

1. Install Hermes Agent

# Clone the repository
git clone https://github.com/NousResearch/hermes-agent.git
cd hermes-agent

# Install dependencies (Python 3.10+)
pip install -e .

# Configure API keys
export ANTHROPIC_API_KEY="your_key_here"
# or use .env file
cp .env.example .env

2. First Conversation

# Start interactive CLI
hermes

# Or one-off command
hermes "analyze this codebase and suggest improvements"

3. Add Your First Skills

Option A: Cross-platform skills (wondelai)

# Clone wondelai/skills (380+ stars, production-ready)
cd ~/.hermes/skills
git clone https://github.com/wondelai/skills.git wondelai

# Skills auto-discovered on next run
hermes "use the youtube search skill to find videos about agent frameworks"

Option B: Literate programming (litprog-skill)

# Clone litprog-skill for code + documentation workflows
cd ~/.hermes/skills
git clone https://github.com/tlehman/litprog-skill.git

# Use literate programming
hermes "create a literate program that implements a binary tree"

4. Add a GUI

Option A: Hermes-native workspace

# Clone hermes-workspace (500+ stars)
git clone https://github.com/outsourc-e/hermes-workspace.git
cd hermes-workspace
npm install
npm run dev
# Access at http://localhost:3000

Option B: Multi-agent orchestration

# Clone mission-control (3.7k+ stars)
git clone https://github.com/builderz-labs/mission-control.git
cd mission-control
docker-compose up -d
# Access at http://localhost:8080

Key Configuration

Profile System (Multi-Instance)

# ~/.hermes/profiles/work.yaml
name: work
messaging_platform: slack
terminal_backend: docker
model: claude-opus-4
memory:
  conversation_window: 50
  auto_summarize: true
# Use specific profile
hermes --profile work "analyze the sales dashboard"

Cron Scheduling

# ~/.hermes/cron.yaml
jobs:
  - name: daily-standup
    schedule: "0 9 * * 1-5"  # 9 AM weekdays
    command: "summarize yesterday's commits and create standup report"
    
  - name: skill-curator
    schedule: "0 2 * * 0"  # 2 AM Sundays
    command: "hermes curator --grade --consolidate --prune"

Messaging Gateway Setup

# Telegram
export TELEGRAM_BOT_TOKEN="your_token"
hermes --messaging telegram

# Discord
export DISCORD_BOT_TOKEN="your_token"
hermes --messaging discord

# Slack
export SLACK_BOT_TOKEN="xoxb-your-token"
export SLACK_APP_TOKEN="xapp-your-token"
hermes --messaging slack

Essential Skills & Plugins

Goal Management & Cost Control

# Install hermes-plugins (42-evey)
cd ~/.hermes/skills
git clone https://github.com/42-evey/hermes-plugins.git

# Usage examples
hermes "set a goal to refactor the authentication module by Friday"
hermes "show me this week's API cost breakdown by model"
hermes "switch to claude-sonnet for this task to save costs"

Inter-Agent Communication

# Using hermes-plugins inter-agent bridge
# Terminal 1: Agent A
hermes --profile researcher --port 8001

# Terminal 2: Agent B
hermes --profile writer --port 8002

# Agent A to Agent B
hermes "send to writer agent: use this research summary to write a blog post"

Incident Response (SRE)

# Install hermes-incident-commander
cd ~/.hermes/skills
git clone https://github.com/Lethe044/hermes-incident-commander.git

# Cron-based monitoring
hermes "monitor production services and alert on failures"

# Manual incident handling
hermes "diagnose why the API is returning 503 errors and apply a fix"

Spotify Control (Linux/Raspberry Pi)

# Install hermes-spotify-skill (Linux-native)
cd ~/.hermes/skills
git clone https://github.com/Alexeyisme/hermes-spotify-skill.git

# Configure Spotify credentials
export SPOTIPY_CLIENT_ID="your_client_id"
export SPOTIPY_CLIENT_SECRET="your_client_secret"
export SPOTIPY_REDIRECT_URI="http://localhost:8888/callback"

# Usage
hermes "play lo-fi hip hop on Spotify"
hermes "skip to the next track"
hermes "transfer playback to my Raspberry Pi speaker"

Self-Hosted Cloud Integration (Nextcloud)

# Install hermes-nextcloud
cd ~/.hermes/skills
git clone https://github.com/adnw-vinc/hermes-nextcloud.git

# Configure
export NEXTCLOUD_URL="https://your-nextcloud.example.com"
export NEXTCLOUD_USERNAME="your_username"
export NEXTCLOUD_APP_PASSWORD="your_app_password"

# Usage
hermes "list my nextcloud files"
hermes "create a note about today's meeting in nextcloud"
hermes "show my calendar for next week"

Deployment Patterns

Production VPS Deployment

# Systemd service
cat > /etc/systemd/system/hermes.service <<EOF
[Unit]
Description=Hermes Agent
After=network.target

[Service]
Type=simple
User=hermes
WorkingDirectory=/opt/hermes-agent
Environment="ANTHROPIC_API_KEY=your_key"
ExecStart=/usr/bin/hermes --messaging telegram --profile production
Restart=always

[Install]
WantedBy=multi-user.target
EOF

sudo systemctl enable hermes
sudo systemctl start hermes

Docker Deployment

# Dockerfile
FROM python:3.11-slim

WORKDIR /app
COPY . .
RUN pip install -e .

ENV ANTHROPIC_API_KEY=""
ENV TELEGRAM_BOT_TOKEN=""

CMD ["hermes", "--messaging", "telegram"]
# Build and run
docker build -t hermes-agent .
docker run -d \
  -e ANTHROPIC_API_KEY="$ANTHROPIC_API_KEY" \
  -e TELEGRAM_BOT_TOKEN="$TELEGRAM_BOT_TOKEN" \
  -v ~/.hermes:/root/.hermes \
  --name hermes \
  hermes-agent

Vercel Sandbox Deployment

# Use Vercel Sandbox backend (serverless)
hermes --terminal vercel-sandbox "deploy this Next.js app"

Modal Deployment (Serverless GPU)

# Use Modal for GPU-intensive tasks
hermes --terminal modal "train this model on the full dataset"

Multi-Agent Orchestration

Using oh-my-hermes Suite

# Install oh-my-hermes (witt3rd)
cd ~/.hermes/skills
git clone https://github.com/witt3rd/oh-my-hermes.git

# Deep research workflow
hermes "research the state of agent frameworks using deep-research skill"

# Consensus planning (Planner → Architect → Critic)
hermes "create a ralplan for building a distributed task queue"

# Verified execution (execute → verify → iterate)
hermes "implement the task queue using ralph skill"

# Full autopilot
hermes "autopilot: research agent frameworks, plan architecture, implement MVP"

Inter-Agent Bridge Pattern

# In agent A's skill
def send_to_agent(target_agent: str, message: str, port: int):
    """Send message to another Hermes instance"""
    import requests
    
    response = requests.post(
        f"http://localhost:{port}/api/message",
        json={"message": message}
    )
    return response.json()

# Usage
send_to_agent("writer", "Use this data to write a report", port=8002)

MCP (Model Context Protocol) Integration

# List available MCP servers
hermes "list mcp servers"

# Use MCP server
hermes "use the filesystem MCP server to read project files"

# Configure custom MCP server
cat > ~/.hermes/mcp-servers.json <<EOF
{
  "servers": {
    "custom-db": {
      "command": "node",
      "args": ["/path/to/db-mcp-server/index.js"],
      "env": {
        "DB_CONNECTION_STRING": "postgresql://..."
      }
    }
  }
}
EOF

Memory System

Conversational Memory

# Search past conversations
hermes "what did we discuss about authentication last week?"

# Summarize conversation history
hermes "summarize all our discussions about the API redesign"

Skill Memory

# Skills automatically improve from experience
# After using a skill multiple times:
hermes "show me how the 'api-integration' skill has evolved"

# Curator grades skills automatically (v0.12.0+)
hermes curator --report

Advanced Patterns

Self-Improving Skill Factory

# Install hermes-skill-factory
cd ~/.hermes/skills
git clone https://github.com/Romanescu11/hermes-skill-factory.git

# Auto-generate skill from workflow
hermes "watch me deploy this app, then create a reusable deployment skill"

Personal Obsidian API

# Install personal-api skill
cd ~/.hermes/skills
git clone https://github.com/beiyuii/personal-api-skill.git

# Point to your Obsidian vault
export OBSIDIAN_VAULT_PATH="$HOME/Documents/Obsidian"

# Usage
hermes "read my daily notes and suggest priorities"
hermes "what did I write about agent frameworks?"

Accumulator Betting Tracker

# Install acca-tracker (sports betting multi-leg tracking)
cd ~/.hermes/skills
git clone https://github.com/svenmedina07-ship-it/skills.git
cd skills/acca-tracker

# Configure cron for 15-minute updates
hermes "set up acca tracking cron job"

# Manual check
hermes "check my accumulator bet status"

API Server Mode

# Start Hermes as API server
hermes serve --port 8080

# Use REST API
curl -X POST http://localhost:8080/api/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "analyze this code", "context": {"file": "main.py"}}'
# Python client
import requests

def ask_hermes(message: str, context: dict = None):
    response = requests.post(
        "http://localhost:8080/api/chat",
        json={"message": message, "context": context or {}}
    )
    return response.json()["response"]

# Usage
result = ask_hermes("refactor this function", context={"file": "utils.py"})

Troubleshooting

Skills Not Loading

# Check skills directory
ls -la ~/.hermes/skills/

# Verify skill structure (needs skill.md or SKILL.md)
cat ~/.hermes/skills/your-skill/skill.md

# Force reload
hermes --reload-skills "list available skills"

Memory Issues

# Clear conversation history
hermes clear-history

# Reduce memory window in profile
# ~/.hermes/profiles/default.yaml
memory:
  conversation_window: 20  # Reduce from default 50

Curator Failures

# Manual curator run with debug
hermes curator --grade --verbose

# Skip consolidation if stuck
hermes curator --grade --prune --skip-consolidate

# Reset curator state
rm ~/.hermes/curator/state.json

Messaging Gateway Issues

# Test connection
hermes --messaging telegram --test-connection

# Debug webhook (for platforms like Telegram)
export HERMES_WEBHOOK_URL="https://your-domain.com/webhook"
hermes --messaging telegram --debug

Terminal Backend Failures

# Test Docker backend
docker ps  # Ensure Docker is running
hermes --terminal docker "echo test"

# Fallback to local execution
hermes --terminal local "echo test"

# SSH backend with custom key
hermes --terminal ssh \
  --ssh-host "192.168.1.100" \
  --ssh-user "ubuntu" \
  --ssh-key "~/.ssh/hermes_key"

Common Workflows

Daily Standup Automation

# ~/.hermes/cron.yaml
jobs:
  - name: daily-standup
    schedule: "0 9 * * 1-5"
    command: |
      summarize yesterday's git commits,
      check open GitHub issues assigned to me,
      create standup report and post to #standup Slack channel

Code Review Assistant

# Review pull request
hermes "review PR #123 on GitHub and provide feedback"

# Continuous code quality
hermes "monitor the main branch and flag potential issues after each push"

Documentation Generator

# Generate docs from code
hermes "analyze src/ and generate API documentation in docs/"

# Keep docs in sync
hermes "watch src/ and update docs/ whenever code changes"

Ecosystem Resources

Maturity Tags

When exploring the ecosystem, note these maturity indicators:

  • production — Stable, documented, actively maintained
  • beta — Works but evolving, expect some rough edges
  • experimental — Proof of concept, learn from it but don't depend on it

Getting Help

# Built-in help
hermes --help

# Skill-specific help
hermes "explain the youtube-search skill"

# Community support
# Post in Discord #hermes-agent channel with:
hermes "generate a debug report" > debug.txt

FAQ

What does the awesome-hermes-agent-ecosystem skill do?

Navigate the Hermes Agent ecosystem — skills, tools, integrations, deployment, and multi-agent orchestration resources

How do I install awesome-hermes-agent-ecosystem?

Run npx skills add https://github.com/reason-machines/ai-agent-skills --skill awesome-hermes-agent-ecosystem in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/awesome-hermes-agent-ecosystem/ (personal) or .claude/skills/awesome-hermes-agent-ecosystem/ (project).

Which AI agents can use awesome-hermes-agent-ecosystem?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is awesome-hermes-agent-ecosystem safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains awesome-hermes-agent-ecosystem?

It is published by reason-machines in reason-machines/ai-agent-skills. It has 176 recorded installs.

More from reason-machines/ai-agent-skills

View all →

agency-agents-zh-skill

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents322 installs

agent-reach-internet-access

reason-machines/ai-agent-skills

Give AI agents eyes to see the internet — scrape Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu with zero API fees

AI & Agents229 installs

agent-sprite-forge-2d-game-assets

reason-machines/ai-agent-skills

Generate 2D sprite sheets, animated GIFs, transparent PNGs, layered maps, and engine-ready prototypes for Godot/Unity using Codex-driven asset pipelines.

Media & Creative217 installs

agentmemory-persistent-memory

reason-machines/ai-agent-skills

Add persistent memory to AI coding agents using agentmemory - remembers context, preferences, and decisions across sessions

AI & Agents216 installs

awesome-agent-skills

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents214 installs

oh-my-openagent-orchestration

reason-machines/ai-agent-skills

Open the skill to read its SKILL.md instructions.

AI & Agents211 installs

Popular ai & agents skills

View category →

google-agents-cli-publish

google/agents-cli

This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).

AI & Agents513.9K installs

google-agents-cli-eval

google/agents-cli

This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).

AI & Agents513.5K installs

skill-creator

anthropics/skills

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

OfficialAI & Agents416.4K installs

codex-pet

prime-skills/runcomfy-agent-skills

Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.

AI & Agents402.1K installs

full-output-enforcement

leonxlnx/taste-skill

Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.

AI & Agents401K installs

executing-plans

obra/superpowers

Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available

AI & Agents245.7K installs