</>MCP Agents Market
Agent

Agentlas-OS

by agentlas-ai1.1kPythonUpdated 2026-08-23

Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model.

Claude CodeCodexGeminiCursorAntigravityOpenCodeDeepSeekOllama

Agentlas-OS (Hephaestus engine) is an open-source agent operating system that creates portable, owner-scoped AI agent packages that work across Claude Code, Codex, Gemini, Cursor, and other LLM platforms. It transforms plain-language requests into runnable single agents or multi-agent teams with explicit contracts, memory boundaries, and verification gates. Users can build agents locally, store them privately in Agent Cloud, or borrow specialists from the public Agentlas Hub. The system operates local-first with model-neutral execution, meaning your chosen LLM and local machine execute the work while packages remain portable across different computers and AI platforms.

Key Features

Portable agent packages that work across Claude Code, Codex, Gemini, Antigravity, Cursor, and local models without rebuilding
Meta-Agent Factory with Briefing Interview Engine that evaluates prompt ambiguity and compiles agents from plain language
Network 2.0 scheduler with federated discovery across local agents, private Agent Cloud, and public Hub specialists
Stormbreaker execution system with deterministic verification gates, parallel work packets, and resumable run journals
Governed memory architecture with local project isolation, curator-gated promotion, and per-agent experience projections
Ontology Runtime with local-first document parsing, CJK trigram search, SQLite storage, and GraphRAG retrieval
Agent-to-Agent (A2A) boundary protocol with caller-gating, redacted exports, and cryptographic import/export
Session-to-Agent builder that converts working conversations into reusable, verified agent packages

Use Cases

  • 01Creating portable customer support agents that can be retrieved and run on any supported LLM platform after signing in
  • 02Building multi-agent teams with explicit roles (orchestrator, worker specialists, QA, memory curator) for complex projects
  • 03Converting successful LLM conversations into reusable agent packages with verified contracts and memory boundaries
  • 04Borrowing public specialist agents from Agentlas Hub for tasks like accessibility QA or market research without exposing local files
  • 05Orchestrating parallel work across frontier and economy models to optimize costs while maintaining quality
  • 06Managing knowledge-intensive operations with local document parsing and semantic search that never sends content to external APIs

Related Agents

View more

hermes-agent

The agent that grows with you

Agent250.4k

agency-agents

A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.

Agent155.5k

openinterpreter

A coding agent for open models like Kimi K3

Agent68k

cline

Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Agent66.3k

Agentlas-OS — FAQ

What is Agentlas-OS?+

Agentlas-OS is an open-source agent operating system (Hephaestus engine) that builds, packages, routes, and executes AI agents across multiple LLM platforms. It creates portable agent packages with explicit contracts, memory rules, and verification gates that work with Claude Code, Codex, Gemini, Cursor, and other supported hosts.

How do I install Agentlas-OS?+

Run the one-line installer script in your terminal: `curl -fsSL https://raw.githubusercontent.com/agentlas-ai/Agentlas-OS/main/scripts/install-all-runtimes.sh | HEPHAESTUS_INSTALL_GLOBAL_ROUTER=1 bash`. This installs the runtime to ~/.agentlas and configures all detected LLM hosts (Claude Code, Codex, Gemini, Cursor, etc.). After installation, restart your AI tool or run `hephaestus doctor` to verify.

Which AI clients does Agentlas-OS work with?+

Agentlas-OS works with Claude Code, Codex, Gemini CLI, Antigravity, Cursor, OpenCode, OpenClaw, Hermes, DeepSeek, GLM, Ollama, and compatible local-model runtimes. The installer automatically configures every supported host it finds on your machine.

Do I need API keys or a subscription?+

Agentlas-OS itself is free and open-source (Apache-2.0 license). You bring your own model accounts or API keys—the system uses whichever LLM and credentials you've already configured in your chosen host. All document processing and semantic search run locally without calling external embedding APIs.

What's the difference between local agents, Agent Cloud, and Agentlas Hub?+

Local agents stay only on your current machine. Agent Cloud is your private, owner-scoped storage where you can save agents and retrieve them on another computer after signing in. Agentlas Hub is the public directory where creators publish specialists that anyone can borrow—borrowed agents don't copy the creator's private source into your workspace.

Is my data sent to external servers?+

Agentlas-OS is local-first: raw documents, prompts, and project files stay on your machine. The Ontology Runtime processes documents locally with SQLite and a bundled embedding model. Agent Cloud stores only agent packages (contracts, routing cards, memory rules), not your credentials, local files, or conversation transcripts.

How do I install Agentlas-OS?+

Open the source repository on GitHub and follow its README. Agentlas-OS is a agent — MCP Agents Market links you directly to the official repo.

Is Agentlas-OS free?+

Agentlas-OS is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches

Developer Tools AgentsProductivity & Workflow AgentsKnowledge & Memory AgentsBest MCP ServersBest SkillsBest PluginsTop rated listingsClaude Code toolsBrowse all categories