MCPInstall

hermes-atlas-ecosystem-map

Build and maintain the Hermes Atlas ecosystem map with quality filtering, RAG chatbot, and live GitHub star tracking

Installs
203
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/reason-machines/hermes-skills --skill hermes-atlas-ecosystem-map
  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 hermes-atlas-ecosystem-map 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.

Hermes Atlas Ecosystem Map

Skill by ara.so — Hermes Skills collection.

What is Hermes Atlas?

Hermes Atlas is a community-curated directory of the Hermes Agent ecosystem—mapping every tool, skill, integration, and deployment template built for Hermes Agent by Nous Research. It features:

  • Quality-filtered repository catalog (80+ projects across 12 categories)
  • Live GitHub star counts with 30-day sparklines
  • RAG-powered chatbot ("Ask the Atlas") grounded in 27 research files
  • Hybrid search/filter with trending badges
  • Zero-framework frontend (vanilla HTML/CSS/JS)
  • Serverless API (Vercel functions + Redis cache)

Installation

git clone https://github.com/ksimback/hermes-ecosystem.git
cd hermes-ecosystem
npm install

Dependencies (only 2):

  • openai — for embeddings and OpenRouter API
  • redis — for star count caching and history

Key Files

hermes-ecosystem/
├── index.html                 # Main map UI (single-page app)
├── data/
│   ├── repos.json            # Single source of truth (84 repos)
│   └── chunks.json           # Pre-computed embeddings (283 chunks, 7MB)
├── api/
│   ├── stars.js              # Live star counts (1hr cache)
│   ├── stars-history.js      # 30-day sparkline data
│   └── chat.js               # RAG chatbot with streaming
├── scripts/
│   ├── build-chunks.js       # Rebuild embeddings from research/
│   └── test-rag.js           # RAG quality tests (27 test cases)
├── research/                  # 27 knowledge base files
└── lib/redis.js              # Shared Redis client

Adding a New Project

1. Quality Filter Criteria

Before adding to data/repos.json, verify:

  • Built for Hermes Agent (not generic AI tools)
  • Created after July 22, 2025 (Hermes Agent launch)
  • Shows genuine effort (not personal pet projects)
  • Passes security review (check for credential leaks, malicious code)

2. Add to repos.json

{
  "categories": {
    "skills": [
      {
        "name": "hermes-skill-example",
        "url": "https://github.com/username/hermes-skill-example",
        "description": "Brief description of what the skill does",
        "tags": ["tag1", "tag2"]
      }
    ]
  }
}

Available categories:

  • skills — Core agent skills
  • tools — External tool integrations
  • plugins — Hermes plugins
  • deployments — Hosting/infrastructure templates
  • frameworks — Multi-skill orchestration
  • integrations — Third-party service connectors
  • forks — Modified Hermes Agent versions
  • ui — User interfaces and dashboards
  • data — Datasets and evaluation tools
  • research — Academic papers and experiments
  • templates — Boilerplates and starters
  • misc — Everything else

3. Verify the Change

# Check JSON syntax
node -e "require('./data/repos.json')"

# Preview locally
open index.html

Rebuilding the Knowledge Base

After updating files in research/, rebuild embeddings:

# Set API key
export OPENROUTER_API_KEY=sk-or-v1-...

# Rebuild chunks.json (splits + embeds)
node scripts/build-chunks.js

# Test RAG quality
node scripts/test-rag.js

build-chunks.js does:

  1. Reads all .md files in research/
  2. Splits into 500-char chunks with 100-char overlap
  3. Embeds each chunk using OpenAI text-embedding-3-small
  4. Writes to data/chunks.json (static file, committed to repo)

Output:

Processing research/agent-architecture.md...
Processing research/hermes-skills.md...
...
✅ Built 283 chunks (7.2 MB)

Testing the RAG Pipeline

export OPENROUTER_API_KEY=sk-or-v1-...
node scripts/test-rag.js

Sample test:

// scripts/test-rag.js excerpt
const tests = [
  {
    query: "How do I create a custom Hermes skill?",
    expectedKeywords: ["skill.md", "yaml", "frontmatter", "triggers"]
  },
  {
    query: "What's the difference between a skill and a tool?",
    expectedKeywords: ["skill", "tool", "integration", "plugin"]
  }
];

Output:

Test 1/27: How do I create a custom Hermes skill?
✅ Found 4/4 keywords in context
...
27/27 passed (100%)

API Endpoints

GET /api/stars

Returns live star counts for all repos in repos.json.

Caching:

  • 1hr TTL in Redis
  • Falls back to direct GitHub API if cache miss
  • Rate limit: 5000/hr with GITHUB_TOKEN, 60/hr without

Response:

{
  "username/repo": 142,
  "anotheruser/hermes-skill": 89
}

Usage in frontend:

const response = await fetch('/api/stars');
const stars = await response.json();
document.querySelector('[data-repo="username/repo"]').textContent = stars['username/repo'];

GET /api/stars-history?repos=user/repo1,user/repo2

Returns 30-day star count history for sparklines.

Response:

{
  "user/repo1": [120, 122, 125, 128, 130, ...],
  "user/repo2": [45, 47, 49, 51, 53, ...]
}

POST /api/chat

RAG chatbot endpoint with streaming support.

Request:

{
  "message": "How do I deploy Hermes Agent?",
  "conversationHistory": [
    {"role": "user", "content": "What is Hermes Agent?"},
    {"role": "assistant", "content": "Hermes Agent is..."}
  ]
}

Response (streaming):

data: {"type":"context","chunks":[{"text":"...","file":"deployment.md"}]}
data: {"type":"token","content":"To deploy"}
data: {"type":"token","content":" Hermes Agent"}
data: {"type":"done"}

Retrieval pipeline:

  1. Conversation-aware rewrite — expands query using chat history
  2. Hybrid search — BM25 (keyword) + cosine similarity (semantic)
  3. MMR re-ranking — maximal marginal relevance to reduce redundancy
  4. Top-K selection — retrieves 5 most relevant chunks
  5. LLM generation — Gemma 4 31B with fallback chain

Configuration

Environment Variables (Vercel)

# Required
OPENROUTER_API_KEY=sk-or-v1-...
REDIS_URL=redis://default:password@host:port

# Optional
GITHUB_TOKEN=ghp_...                              # Fine-grained PAT (public read-only)
OPENROUTER_MODEL=google/gemma-4-31b-it:free       # Primary LLM
OPENROUTER_FALLBACK_MODELS=google/gemma-4-26b-it:free,google/gemini-3-flash-1.5

vercel.json Configuration

{
  "functions": {
    "api/**/*.js": {
      "maxDuration": 30
    }
  },
  "crons": [
    {
      "path": "/api/stars-daily-snapshot",
      "schedule": "0 0 * * *"
    }
  ]
}

Common Patterns

Adding Multiple Projects at Once

// scripts/batch-add.js (create this if needed)
const fs = require('fs');
const repos = require('../data/repos.json');

const newProjects = [
  { name: "hermes-skill-web-search", url: "https://github.com/...", category: "skills" },
  { name: "hermes-tool-calendar", url: "https://github.com/...", category: "tools" }
];

newProjects.forEach(proj => {
  repos.categories[proj.category].push({
    name: proj.name,
    url: proj.url,
    description: "",  // Fill in manually
    tags: []
  });
});

fs.writeFileSync('./data/repos.json', JSON.stringify(repos, null, 2));

Security Review Checklist

// scripts/security-check.js (conceptual)
const checks = [
  {
    name: "No hardcoded credentials",
    test: (code) => !/api[_-]?key\s*=\s*["'][^"']+["']/i.test(code)
  },
  {
    name: "No eval() usage",
    test: (code) => !/eval\(/.test(code)
  },
  {
    name: "Dependencies up to date",
    test: async (repo) => {
      // Check package.json for known vulnerable versions
    }
  }
];

Custom Sparkline Rendering

// From index.html (adapted for skill documentation)
function renderSparkline(history) {
  const max = Math.max(...history);
  const min = Math.min(...history);
  const range = max - min || 1;
  
  const points = history.map((val, i) => {
    const x = (i / (history.length - 1)) * 100;
    const y = 100 - ((val - min) / range) * 100;
    return `${x},${y}`;
  }).join(' ');
  
  return `<svg viewBox="0 0 100 100"><polyline points="${points}" /></svg>`;
}

Troubleshooting

Embeddings Build Fails

Error: Error: 429 Too Many Requests

Fix: Add rate limiting to build-chunks.js:

async function sleep(ms) {
  return new Promise(resolve => setTimeout(resolve, ms));
}

for (const chunk of chunks) {
  const embedding = await getEmbedding(chunk.text);
  chunk.embedding = embedding;
  await sleep(100); // 10 req/sec max
}

Redis Connection Timeout

Error: ECONNREFUSED or ETIMEDOUT

Fix: Check REDIS_URL format:

# Correct format
redis://default:password@host.region.cloud.redislabs.com:12345

# Common mistake (missing default user)
redis://:password@host...

Stars Not Updating

Error: Cached star counts stale after 1 hour

Fix: Manually bust cache:

# Via Redis CLI
redis-cli -u $REDIS_URL
> DEL stars:cache

Or programmatically:

// api/stars.js — force refresh
if (req.query.refresh === 'true') {
  await redis.del('stars:cache');
}

RAG Returns Irrelevant Results

Symptom: Chatbot answers unrelated to Hermes Agent

Fix: Check chunk quality:

node scripts/test-rag.js --verbose

# Review chunks.json for short/low-quality chunks
node -e "
const chunks = require('./data/chunks.json');
const short = chunks.filter(c => c.text.length < 200);
console.log('Short chunks:', short.length);
"

Increase chunk size in build-chunks.js:

const CHUNK_SIZE = 800;  // was 500
const OVERLAP = 150;     // was 100

GitHub Rate Limit Exceeded

Error: 403 rate limit exceeded

Fix: Add GITHUB_TOKEN to Vercel env vars:

  1. Go to GitHub → Settings → Developer settings → Personal access tokens → Fine-grained tokens
  2. Create token with Public repositories (read-only) permission
  3. Add to Vercel: GITHUB_TOKEN=ghp_...

This increases rate limit from 60/hr to 5000/hr.

Local Development

# Open the map in a browser (API endpoints won't work locally)
open index.html

# Test API endpoints locally (requires env vars)
export OPENROUTER_API_KEY=...
export REDIS_URL=...
vercel dev

Note: The frontend is pure static HTML—no build step required. API endpoints only run on Vercel or with vercel dev.

Deployment

# Deploy to Vercel
vercel --prod

# Set environment variables
vercel env add OPENROUTER_API_KEY
vercel env add REDIS_URL
vercel env add GITHUB_TOKEN

The daily cron job (/api/stars-daily-snapshot) automatically runs at midnight UTC to populate sparkline history.


Live site: hermesatlas.com
Repository: github.com/ksimback/hermes-ecosystem

FAQ

What does the hermes-atlas-ecosystem-map skill do?

Build and maintain the Hermes Atlas ecosystem map with quality filtering, RAG chatbot, and live GitHub star tracking

How do I install hermes-atlas-ecosystem-map?

Run npx skills add https://github.com/reason-machines/hermes-skills --skill hermes-atlas-ecosystem-map 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/hermes-atlas-ecosystem-map/ (personal) or .claude/skills/hermes-atlas-ecosystem-map/ (project).

Which AI agents can use hermes-atlas-ecosystem-map?

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 hermes-atlas-ecosystem-map 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 hermes-atlas-ecosystem-map?

It is published by reason-machines in reason-machines/hermes-skills. It has 203 recorded installs.

More from reason-machines/hermes-skills

View all →

hermes-agent-self-evolution

reason-machines/hermes-skills

Evolutionary self-improvement for Hermes Agent using DSPy + GEPA to optimize skills, prompts, and code

AI & Agents323 installs

hermes-webui-agent

reason-machines/hermes-skills

Expert in deploying, configuring, and using Hermes WebUI—a web interface for Hermes Agent with persistent memory, scheduled jobs, and multi-platform messaging.

267 installs

awesome-hermes-agent-ecosystem

reason-machines/hermes-skills

Expert knowledge of the Hermes Agent ecosystem, skills, plugins, tools, integrations, and deployment patterns for building and extending AI agents

AI & Agents239 installs

hermes-agent-framework

reason-machines/hermes-skills

Hermes Agent framework by Nous Research - self-improving AI agent with built-in learning loop, three-layer memory, and automatic skill evolution

AI & Agents222 installs

hermes-workspace-ai-agent-ui

reason-machines/hermes-skills

Native web workspace for Hermes Agent with chat, terminal, memory, skills, inspector, and multi-agent orchestration

AI & Agents218 installs

hermes-web-ui-dashboard

reason-machines/hermes-skills

Web dashboard for managing Hermes Agent multi-platform AI chat sessions, analytics, scheduled jobs, and platform channels

Data & Databases210 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