{"slug": "agentos-a-rust-runtime-for-ai-agents-with-deterministic-time-travel-replay", "title": "AgentOS: a Rust runtime for AI agents with deterministic time-travel replay", "summary": "Developer Wahib El Khadiri has released AgentOS, an open-source Rust runtime for AI agents that provides deterministic time-travel replay. The runtime journals every LLM exchange and tool result, enabling offline replay and forking into alternate timelines without API costs. AgentOS sits underneath frameworks like LangGraph and AutoGen, offering supervision, health endpoints, and a debugging dashboard.", "body_md": "Most agent frameworks help you *build* a workflow. The harder part starts after that: the workflow has to run as a long-lived process, fail clearly, restart carefully, and be inspectable after the fact.\n\nThat's the gap I'm building **AgentOS** for — an open-source, Rust-first runtime layer that sits *underneath* frameworks like LangGraph, AutoGen or CrewAI instead of replacing them.\n\n```\ncargo run -p agentos-cli -- run --agent examples/simple_agent.toml\n```\n\nThat single command brings up a supervised agent, a health endpoint, a gRPC message bus, a live SSE event stream, and a recorded trace you can replay later. No API key is needed just to bring the runtime up.\n\nYour agent does something weird on step 7. Reproducing it costs real API calls, and it never behaves the same way twice.\n\nAgentOS journals every LLM exchange and tool result at the provider boundary, so any run can be replayed deterministically — and forked into alternate timelines:\n\n```\nagentOS run --agent my_agent.toml    # every step journaled automatically\nagentOS replay --session agent_123   # offline re-run, no API cost, drift-checked\nagentOS fork --from ckpt_4 --prompt \"try the other path\"\n```\n\nThe dashboard's Recordings view turns those journals into a scrubbable timeline: step through the prompt, each exchange, tool calls and their results, with per-exchange checkpoints as fork anchors.\n\n`crates/kernel`\n\n— lifecycle, agent handles, supervisor`crates/bus`\n\n— in-memory, gRPC, SSE and WebSocket messaging`crates/trace`\n\n— recording, replay, diff, checkpoint model`crates/vault`\n\n— secret isolation, encryption, scopes, audit`crates/memory`\n\n, `crates/registry`\n\n, `crates/llm`\n\n, `crates/cli`\n\n, `crates/sdk`\n\n`dashboard/`\n\n— React debugging surfaceStable enough for local use: the `run / ps / logs / trace / replay`\n\nCLI flows, local state inspection, export and import, and the core crates with workspace checks and tests.\n\nStill experimental: the dashboard, the WASM plugin runtime, Docker Compose packaging, LLM provider integrations, and Python/TypeScript SDK packaging. Stronger restart and recovery guarantees with explicit tests are next. It is an alpha, and I would rather say that than claim production hardening it has not earned yet.\n\n```\ngit clone https://github.com/WAHIB-EL-KHADIRI/AgentOS\ncd AgentOS\ncargo build --workspace\ncargo run -p agentos-cli -- run --agent examples/simple_agent.toml\n```\n\nRepo: [https://github.com/WAHIB-EL-KHADIRI/AgentOS](https://github.com/WAHIB-EL-KHADIRI/AgentOS)\n\nFeedback on the runtime boundaries is what I want most: if you run agents today, what breaks first for you — supervision, observability, or reproducing failures?", "url": "https://wpnews.pro/news/agentos-a-rust-runtime-for-ai-agents-with-deterministic-time-travel-replay", "canonical_source": "https://dev.to/wahib_el_khadiri_0/agentos-a-rust-runtime-for-ai-agents-with-deterministic-time-travel-replay-bmm", "published_at": "2026-07-26 18:22:50+00:00", "updated_at": "2026-07-26 19:01:11.862133+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools"], "entities": ["AgentOS", "Wahib El Khadiri", "LangGraph", "AutoGen", "CrewAI", "Rust"], "alternates": {"html": "https://wpnews.pro/news/agentos-a-rust-runtime-for-ai-agents-with-deterministic-time-travel-replay", "markdown": "https://wpnews.pro/news/agentos-a-rust-runtime-for-ai-agents-with-deterministic-time-travel-replay.md", "text": "https://wpnews.pro/news/agentos-a-rust-runtime-for-ai-agents-with-deterministic-time-travel-replay.txt", "jsonld": "https://wpnews.pro/news/agentos-a-rust-runtime-for-ai-agents-with-deterministic-time-travel-replay.jsonld"}}