Agents that don't just answer once, but keep working toward a goal: understand the objective, build a trackable task graph, call governed capabilities, keep the evidence, resume after waiting, failure or disconnection — and explain the whole thing afterwards.
Web, CLI, SDK, enterprise apps and edge executors don't each implement their own agent. They share the same session, run, Work, context and audit semantics — they differ only in what capabilities they can reach.
Shared runtime backbone
Cloud capabilities
Sessions, memory, tasks, reports and governed external access
Cloud servicesEdge executor
Local files, shell, Git, browser and private networks
User device / workstationMCP capabilities
Knowledge bases, databases, ticketing and approval systems
MCP service environmentSandbox executor
Workloads needing isolation, quotas or a specific runtime
Sandbox runtimeClients are different entrances, providers are different hands — the thinking, the state and the recovery always live in one backbone.
Continuity
Work doesn't depend on a page or a process. It pauses, resumes and follows you across devices.
Control
Capability, permission, network, data and side-effect boundaries are explicit — block one action precisely without killing the whole task.
Explainability
Not just the answer: which context was used, which tools were called, what failed along the way, and why this route was chosen.
The same problem can be met by tuning prompts from experience, or treated as a systems problem that can be optimized and measured. Astra takes the second path.
Peng Xu, Zuyu Zhang, Yuze Sun, Feng Tian, Long Wang, Chen Zhang
Every model call in a long-horizon agent has to answer the same question: what goes into this prompt, in what order, and when does history get compacted. And it has to answer under two hard constraints — a fixed context-window budget, and a byte-sensitive prompt cache where changing one byte can invalidate the whole segment. The paper brings database query execution to that problem: data sources are catalogued like tables, assembly runs as a five-phase pipeline, and the optimization is deterministic rather than left to the model.
Measured on the Qutebrowser subset of SWE-bench Pro, against append-only context assembly.
Read the paperThis is the line between an agent platform and a chat endpoint.
One question usually takes one model call. Real work doesn't: analysing code, debugging an incident, working a ticket queue, running a deployment, producing a report — all of it takes many steps and many tools. Once work gets long, the hard parts show up. One HTTP request can't cover the whole execution. Should the task continue after the UI disconnects? Retry, reroute or stop when a tool fails? How does a human approval get back into the original task? How does this turn's evidence reach the next one? Astra makes that state explicit and durable — part of the runtime, instead of scattered across UI, scripts and prompts.
“Long-running work doesn't need a smarter prompt. It needs an explicit, durable, resumable runtime.”
Work and the task graph make a long goal visible: what's done, what's running, what's waiting. You always know what the agent is doing.
Every step's output and evidence is kept, so the next turn can rely on it instead of asking the model to remember.
Runs checkpoint at meaningful boundaries. After a failure, a wait or a disconnect, work continues from there — not from scratch.
A run is a controllable execution: cancel it, pause it, or let it continue once a condition is met. The boundaries are explicit.
When a human is needed the run enters an explicit wait, and the answer returns to the original task rather than starting a new thread.
Work keeps moving after the connection drops, then delivers a result or a failure reason — with traceable evidence and a next step.
The cloud is the right place for shared state, model calls and orchestration. Your files, code, browser sessions and private network usually are not. The edge executor lets an agent work in your workspace without giving up your local data boundary.
Cloud runtime backbone
Shared facts and coordination
Edge executor
Hands inside an authorised boundary
The edge executor does not run a second agent loop. The cloud holds the control facts for sessions, runs and Work; the edge only executes local work inside a bound, authorised boundary.
What happens when the link drops
A side effect isn't judged by the tool's name, but by what this call really changes, who it affects, whether it can be rolled back, and whether it crosses a trust boundary. The runtime then decides: run it, ask first, allow it only as a suggestion in planning mode, or block it.
| Tier | Examples | Typical policy |
|---|---|---|
| Read-only | Read a file, check status, query a report | Allowed if provider and user permissions permit |
| Local change | Write files, run tests, modify Git | Requires a bound workspace and approval |
| External read | Browse the web, call a business API | Subject to network and data policy |
| External change | File a ticket, publish, write to a database | Explicit approval or a higher trust level |
| High risk | Deletion, credential exposure, dangerous network ops | Denied by default or hard-gated |
Planning mode: think first, then act
Planning mode lets the agent understand the problem, read context, break down steps and propose a plan — while side effects like writing files, changing code or sending external messages stay blocked by default. It's a change of execution policy, not a second agent.
Frameworks give you parts. Runtimes give you guarantees.
| Dimension | Framework (LangChain / CrewAI) | Coding Assistant (Cursor / Claude Code) | Astra Agentic Runtime |
|---|---|---|---|
| Form | A pile of libraries, DIY | Single-purpose product | Managed runtime + guarantees |
| What you write | Wire everything from scratch | Use someone else's tool | Just a system prompt + skills |
| Decision audit | None | None | Every decision bound to a snapshot |
| Memory management | RAG bolted on | None | 5-layer cognitive architecture, self-curating |
| Production experiments | Gamble in prod | N/A | Clone prod DB, zero-risk regression |
| Cost & trust | Each piece separate | Black box | Unified cost + confidence signals |
“JVM is to Java what Astra is to Agents.”
Agent developers write prompts and pick skills — the runtime does the rest.
Edge runs tools, cloud owns LLM and state, MatrixOne powers enhanced services.
Adding a new business agent = define a system prompt + pick skills + pick a model. Zero platform code changes.
Three steps that reveal Astra's runtime guarantees.
Walk the causal chain → hybrid-search the memory → load skill definitions → allocate token budget by task type. Every token has to earn its place.
Before the LLM call, asynchronously write the context snapshot. Model, parameters and input are all frozen for later audit and replay.
Hallucination Firewall verifies each claim → decision lands in audit log → implicit feedback flows back → regression gate drives the next round of prompt and skill evolution.
Every step written into the event stream and snapshots — ask 'why did the agent do that' three months later, and get the exact answer in seconds.
From customer-facing assistants to platform-internal guardian agents.
Every reply traces back to the product doc version cited at the time. When docs update, impacted past answers auto-enter QA review — no more serving stale information to customers.
The agent generates recommendations from customer history, contract libraries and compliance rules. Each suggestion carries a confidence score so reviewers can focus only on low-confidence sections — faster without losing control.
Business users ask questions in natural language. The agent runs analysis on a zero-copy clone of production — heavy queries don't impact live workloads. Results bind to data snapshots so you can recreate the exact state later.
When recommending segmentation or campaign strategies, the agent runs four approaches in parallel in sandboxed clones, computes expected ROI from historical data, then suggests the best. Decisions are grounded and reviewable.
Every suggestion binds to the code snapshot and repo state at the time. Three months later, you can reconstruct the full code context of any past suggestion — no more 'the AI said something but we can't reproduce it.'
Runs as a System Agent triggered by failures: pull context → reproduce in a cloned environment → propose a fix. Every diagnostic step is recorded — on-call has receipts.
Rust-first performance core, MatrixOne as platform state store.
A capability knowledge base and RFQ parser let an Agent classify each requirement as compliant, non-compliant, or uncertain, cite evidence, and fill the original Excel template.
A legal team at a multi-billion-dollar multinational handles a large volume of cross-border contracts, with long manual review cycles, risk findings that could not be traced, and standards that varied by reviewer. Amway built a unified legal knowledge brain on MOI that flags risk clauses, missing clauses and inconsistent definitions in seconds.
Industry scanning, target identification and post-investment tracking all ran on manual effort, with internal and external data siloed apart. The fund built an AI investment intelligence system on MOI: multi-source data is collected and structured automatically, linked into industry-chain, company and technology knowledge graphs, with reports and alerts generated on top.
Students don't know what to read or how; parents can't see progress; teachers can't follow every child. Tencent built AI Tutor on MOI, using a multi-agent architecture that turns one reading session into a personalised path, interactive Q&A, comprehension testing and knowledge expansion.
MOI connects procurement document parsing, item confirmation, multi-source research, supplier recommendations, price analysis, and report export in one evidence-backed workflow.
A leading new energy manufacturer built an AI data platform with MatrixOne Intelligence, enabling end-to-end automated bid document generation and improving bidding decision efficiency by 40%.
Astra already powers multiple business agents running in production inside MatrixOrigin. Contact us for a detailed solution or book a deep-dive demo.