agentlas-ai/Agentlas-OS
Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model. observed · 2026-09-03
Health v2 · maintenance only
77/100
- Activity 100
- Release rhythm 88
- Longevity 6
Flags: young
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: 0.0
- age_days: 90
- days_rel: 1
- days_push: 0
- n_releases_24m: 211
Adoption not part of the score
1099 stars · 103 forks observed · 2026-09-03
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded
Agentlas OS (with the Hephaestus engine) is an open-source agent framework that lets users create portable AI agents from plain-language requests, keep them in a local hub, and spin up temporary orchestrators per task. It works model-agnostically across runtimes like Claude Code, Codex, Gemini, Cursor, and Ollama, with an optional Hub for borrowing specialist agents and a private Agent Cloud for owner-scoped storage.
Use cases
- build ai agents from a plain-language prompt
- orchestrate a team of specialist agents for a task
- run the same agent across claude code, codex, and gemini
- borrow prebuilt specialist agents from a public hub
- keep my agents portable and owner-scoped in a private cloud
- set up multi-agent workflows locally without vendor lock-in
- install an agent framework as a plugin into my existing llm cli
- monitor and chat with my agents from my phone
When to choose
- you want model-agnostic, local-first multi-agent orchestration
- you want agents as portable assets that outlive a single tool or machine
- you want to install agent capabilities into existing LLM CLIs like Claude Code or Codex
- you want a hub economy to borrow or publish specialist agents
- you prefer Apache-2.0 open-source tooling with a desktop app
When to avoid
- you need a battle-tested enterprise orchestration platform with long production history
- you want a fully self-contained system with no hosted Hub/Cloud dependency or credits metering
- you only need simple deterministic workflow automation rather than autonomous agents
- you are uncomfortable with an installer script that pipes from the internet into your LLM runtime config
Facets
framework · maturity active
agent-framework mcp llm-inference workflow-automation plugin-system chatbot large-language-models developer-tools self-hosted windows python cross-platform cli multi-agent-orchestration a2a agent-hub agent-marketplace local-first agent-portability orchestrator llm-runtime-agnostic desktop-app agent-cloud ai-agents automation macos linux desktop
10 sources
- readme: https://github.com/agentlas-ai/Agentlas-OS · fetched 2026-09-03 · 462543b3cac5
- homepage: https://agentlas.cloud · fetched 2026-08-29 · b1edcc9e0716
- site_page: https://agentlas.cloud/docs · fetched 2026-08-29 · 18810b745d33
- site_page: https://agentlas.cloud/docs/introduction/agentlas-one · fetched 2026-08-29 · 4d17c43a68e2
- site_page: https://agentlas.cloud/docs/guides/build · fetched 2026-08-29 · 4ad4f859ab61
- site_page: https://agentlas.cloud/docs/changelog · fetched 2026-08-29 · c636cdb5c00e
- site_page: https://agentlas.cloud/docs/legal/terms · fetched 2026-08-29 · e9ec90f5a193
- site_page: https://agentlas.cloud/docs/legal/privacy · fetched 2026-08-29 · b3135b642b52
- site_page: https://agentlas.cloud/pricing · fetched 2026-08-29 · e3e2de94cd4b
- site_page: https://agentlas.cloud/updates/automation-tools-are-not-ai · fetched 2026-08-29 · 1b6b3372af6e
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| agentlas-ai/Agentlas-OS | main | 77 |
For agents
markdown · JSON · MCP: product_card(name="agentlas-ai/Agentlas-OS")
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem