ATOM Platform

Open-Source AI Agent Platform for Self-Hosted Automation

Developer Note: For technical setup and architecture, see docs/development/overview.md.

Atom Platform

Automate your workflows by talking to an AI — and let it remember, search, and handle tasks like a real assistant.

</div>

What is Atom?

Atom is an open-source, self-hosted AI agent platform that combines visual workflow builders with intelligent LLM-based agents.

Just speak or type your request, and Atom's specialty agents will plan, verify, and execute complex workflows across your entire tech stack.

Key Difference: Atom is self-hosted — your workflow data, agent state, and memory stay on your infrastructure. LLM inference uses your own API keys (BYOK) with cloud providers (OpenAI, Anthropic, DeepSeek); local model support (Ollama, Llama.cpp) is available for fully private deployments.

Comparing alternatives? See Atom vs OpenClaw for a detailed feature comparison.

Atom vs OpenClaw: Quick Comparison

Aspect	Atom	OpenClaw
Best For	Business automation, multi-agent workflows	Personal productivity, messaging workflows
Agent Model	Multi-agent system with specialty agents	Single-agent runtime
Governance	✅ 4-tier maturity (Student → Autonomous)	❌ No maturity levels
Memory	✅ Episodic memory + per-turn fact extraction + agent memory tools	✅ Persistent Markdown files
Integrations	46+ business (CRM, support, dev tools)	50+ personal (smart home, media, messaging)
Architecture	Python + FastAPI + PostgreSQL/SQLite	Node.js + local filesystem
Setup	Docker Compose (~15-30 min)	Single script (~10-30 min)

Full Comparison →

Atom vs Hermes Agent: Quick Comparison

Hermes (Nous Research) is an open-source personal agent known for its memory-provider architecture. Atom adopted its strongest ideas (per-turn fact extraction, pre-compression hooks, circuit breaker, FTS5 search) and deliberately avoided its weakest (custom LLM-summarizing compressor — Hermes' own has 3 documented production bugs).

Aspect	Atom	Hermes Agent
Best For	Business automation, governed multi-agent workflows	Personal coding/productivity assistant
Memory extraction	✅ Per-turn durable-fact extraction (5 categories, Mem0 taxonomy)	✅ Memory-provider ABC with 7 hooks (reference design)
Context compression	✅ Boundary protection + tool-pair sanitization (deterministic only)	◐ 4-phase compressor incl. LLM summary (3 documented bugs)
Agent memory tools	✅ `memory_remember` / `memory_forget` (maturity-gated)	✅ `lancedb_remember` / `mem0_*` tool family
Governance	✅ 4-tier maturity (Student → Autonomous) + HITL supervision	❌ None
Multi-agent	✅ Queen + Fleet Admiral + specialty agents	❌ Single agent loop
Canvas / rich UI	✅ 7 canvas types, WebSocket, a11y	❌ Terminal + messaging
Cost routing	✅ 5-tier cognitive classification	◐ Aux-model only
Observability	✅ Prometheus + `/health/*` + structlog	❌ WARNING logs
Retrieval	Tier-1 SQL + Tier-2 vector + FTS5 lexical	Vector + BM25 hybrid + cross-encoder reranker
Circuit breaker	✅ 5 failures → 120s cooldown → half-open probe	✅ 2-min/5-failure (fixed window)
Deployment	Python + FastAPI + SQLite/PostgreSQL + embedded LanceDB	Python self-hosted + embedded LanceDB

Full Comparison → · Context Memory Design →

Architecture

Single-Tenant Deployment

Atom is designed for self-hosted deployment:

Simpler Setup: No tenant isolation, no subdomain routing
Better Performance: Direct database access without overhead
Self-Hosted: Agent state, memory, and workflow data stay on your infrastructure. LLM prompts are sent to your configured API provider (BYOK). Use local models (Ollama/Llama.cpp) for fully private setups.
Unlimited Usage: No subscription fees or quota limits

Key Features:

Uses user_id for user identification
No billing system or quota enforcement
Fleet recruitment limited by system resources only
All governance, routing, and graduation features work identically

Full Architecture Guide →

Data Flow & Privacy

Understanding where your data goes:

Component	Where Data Goes	Configurable?
LLM inference (chat, reasoning, agent decisions)	Cloud API provider via your BYOK keys (OpenAI, Anthropic, DeepSeek)	✅ Use local models (Ollama, Llama.cpp) for fully private
Embeddings (document vectors)	Local (FastEmbed, ONNX runtime)	Always local
Vector storage (episodic memory)	Local (LanceDB on disk)	Always local
Database (agents, users, workflows)	Local (SQLite) or your PostgreSQL server	Always your infra
File uploads	Local filesystem (`./data/`)	Always your infra
Integration data (Slack, Gmail, etc.)	Third-party APIs per integration	Per-integration

For maximum privacy: Set ATOM_LOCAL_ONLY=true (blocks cloud integrations) AND configure local LLM models (Ollama/Llama.cpp) instead of cloud API keys.

Meta-Agent Routing ✨

Intelligent CHAT/WORKFLOW/TASK routing with governance checks and dynamic fleet recruitment

Meta-Agent Guide →

Quick Start

bash
# Install Atom
pip install atom-os

# Initialize
atom init

# Add your API keys to .env
# OPENAI_API_KEY=sk-...
# MINIMAX_API_KEY=...  (optional, for MiniMax M2.7 support)

# (Optional) Connect to marketplace for commercial specialty skills
# MARKETPLACE_API_TOKEN=at_saas_your_token  # Get from https://atomagentos.com


# Start Atom
atom start

# Open dashboard
open http://localhost:8000

That's it! 🚀

Optional: Connect to Marketplace →

Choose your edition:

Personal Edition - Free, single-user, SQLite (default)
Enterprise Edition - Multi-user, PostgreSQL, monitoring → pip install atom-os[enterprise]

Full Installation Guide →

Key Features

🎙️ Voice Interface

Build complex workflows using just your voice
Natural language understanding — no proprietary syntax
Real-time feedback as Atom visualizes its reasoning

🤖 Specialty Agents & Orchestration ✨ Enhanced 2026

Sales, Marketing, Engineering: CRM pipelines, campaigns, deployments, incidents
Hive Orchestration: Queen Agent (structured workflows) and FleetAdmiral (dynamic recruitment)
Conductor Agent: 5 execution strategies (SEQUENTIAL, PARALLEL, HYBRID, ADAPTIVE, ROLLBACK_SAFE)
Workflow State Machine: Validated transitions with automatic rollback
Event Bus: Event-driven workflow triggering with pub/sub
Self-Evolving Capabilities: Memento Skills learns from failures, AlphaEvolver optimizes via mutation

Special Agents Guide → | Queen Agent → | Auto-Dev →

🎨 Canvas Presentations & Real-Time Guidance ✨

Rich interactive presentations (charts, forms, markdown) with live operation visibility, multi-view orchestration, smart error resolution, and AI accessibility (canvas state exposed to agents)

Canvas Guide →

🧠 Autonomous Self-Evolution & Graduation ✨

Experience-based learning with recursive self-evolution, dual-trigger graduation (SUPERVISED → AUTONOMOUS), and hybrid PostgreSQL + LanceDB storage

Agent Graduation Guide →

💾 Memory & Context (Hermes-style) ✨ New 2026

Durable-fact extraction layer that survives context compression — the agent remembers what matters across sessions:

Per-turn extraction: 5 durable-fact categories (exact values, hard constraints, decision reasoning, cross-task deps, implicit preferences) extracted fire-and-forget after each ReAct step
Two-tier recall: Tier-1 pure-SQL DURABLE FACTS prompt block (sub-ms) + Tier-2 LanceDB semantic recall (opt-in) + FTS5 lexical fallback for exact-match queries
Agent memory tools: memory_remember / memory_forget let the agent explicitly persist or invalidate facts mid-turn (maturity-gated, deletion-safe)
Pre-compression queue: drains prompts before truncation drops facts (strictly additive, default ON)
Circuit breaker: 5 failures → 120s cooldown → half-open probe (prevents extraction storms)
Boundary-protection compression: head + tail preserved, stale middle elided (deterministic; no buggy LLM-summary phase)

Context Memory Design → · Atom vs. Hermes →

🛡️ Agent Governance System ✨ Enhanced 2026

4-tier maturity: Student → Intern → Supervised → Autonomous
Three-layer governance: OPERATIONAL (<10ms), TACTICAL (<100ms), STRATEGIC (human-in-the-loop)
Policy engine: Context-aware evaluation with priority resolution
AI-powered training: Duration estimation with historical data
Complete audit trail: Every action logged, timestamped, and traceable

Governance Documentation → | Enhancement Plan →

🔌 Deep Integrations

46+ business integrations: Slack, Gmail, HubSpot, Salesforce, Zendesk
9 messaging platforms: Real-time communication
Marketplace Connection: Access 5,000+ community skills and agent templates ✨ NEW
Use /run, /workflow, /agents from your favorite chat app

🔍 Knowledge Graph & GraphRAG ✨ Enhanced 2026

Recursive knowledge retrieval via BFS traversal, canonical anchoring to database records, bidirectional sync, and D3-powered visual explorer

Multi-Hop Expansion: Cue-driven activation for entity relationships
Dynamic Graph Construction: Incremental updates without full rebuilds
Enhanced Community Detection: Leiden algorithm clustering

GraphRAG Documentation →

🌐 Community Skills & Package Marketplace ✨

5,000+ OpenClaw/ClawHub skills with PostgreSQL marketplace, LLM-powered security scanning (21+ malicious patterns), DAG skill composition, Python + npm auto-installation with vulnerability scanning, and supply chain protection

Community Skills Guide → | Python Packages → | npm Packages →

🔍 Browser & Device Automation

Browser automation via CDP (scraping, form filling)
Device control (camera, location, notifications)
Maturity-governed for security

🚀 2026 Enhancement Plan

Based on cutting-edge 2025-2026 AI research, Atom has been enhanced with 5 major feature phases:

Phase 1: Memory & Graduation ✅ Complete

POMDP memory framework for experience-driven learning
Offline memory consolidation (inspired by human sleep)
Quality-weighted graduation criteria (20% improvement)

Phase 2: GraphRAG Enhancement ✅ Complete

Multi-hop expansion with cue-driven activation
Dynamic graph construction (incremental updates)
Enhanced community detection (Leiden algorithm)

Phase 3: Learning-Based LLM Routing ✅ Complete

RouteLLM training from user preferences
Predictive cache warming and optimization
Additional 15% cost reduction on top of cache savings

Phase 4: Zero-Trust Federation Identity ✅ Complete

DID (Decentralized Identifiers) for cryptographic identity
Verifiable Credentials (VCs) for signed claims
Zero-trust security framework with per-request verification
Automatic credential rotation (90-day)

Phase 5: Enhanced Orchestration Patterns ✅ Complete

Conductor Agent (5 execution strategies)
Workflow State Machine (validated transitions with rollback)
Event Bus (pub/sub event-driven triggering)
Workflow Templates & Composition (8 primitives)

Performance: 193 tests passing, comprehensive validation metrics documented

Enhancement Plan → | Validation Metrics →

Installation Options

🐳 Docker (5 minutes)

bash
git clone https://github.com/rush86999/atom.git
cd atom
cp .env.personal .env
docker-compose -f docker-compose-personal.yml up -d

🚀 DigitalOcean (1-Click Deploy)

Launch Atom on DigitalOcean App Platform with one click:

See Cloud Deployment Guide →

💻 Native (10 minutes)

bash
git clone https://github.com/rush86999/atom.git
cd atom
cd backend && python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
cd ../frontend-nextjs && npm install
cp .env.personal .env
# Start backend: cd backend && python -m uvicorn main_api_app:app --reload
# Start frontend: cd frontend-nextjs && npm run dev

Marketplace (Commercial Service)

Commercial marketplace for agents, domains, components, and skills at atomagentos.com. Requires API token connection. Core platform is AGPL v3 (open source), marketplace items are proprietary. See LICENSE.md for terms.

Setup: Add ATOM_SAAS_API_TOKEN to .env and restart. Marketplace Documentation →

Example Use Cases

Department	Scenario
Sales	New lead in HubSpot → Research → Score → Notify Slack
Finance	PDF invoice in Gmail → Extract → Match QuickBooks → Flag discrepancies
Support	Zendesk ticket → Analyze sentiment → Route urgent → Draft response
HR	New employee in BambooHR → Provision → Invite → Schedule orientation

Security & Privacy

Self-hosted deployment, BYOK (OpenAI/Anthropic/Gemini/DeepSeek/MiniMax), encrypted storage (Fernet), audit logs, human-in-the-loop approvals, package security scanning, supply chain protection, comprehensive testing (495+ tests), AI-enhanced bug discovery, and stress testing

Security Documentation → | Testing Guide →