Prototyping an agent = easy. Running it reliably at scale = the real challenge. We'll build one live @AISummitSeoul — stateful workflows, secure tool execution, sandboxed runtime — with LangGraph + CubeSandbox on OCI Bare Metal. 📅 Aug 19–20 · COEX Grand Ballroom & Exhibition Hall B 👉 @CubeSandbox #TencentCloud# #CubeSandbox# #Oracle# #OCI# #LangGraph# #AISummitSeoul# #AIAgents# #AgentInfrastructure#

🧩 DeepSeek and OpenAI Open-Sourced Their Harnesses. The Runtime May Become the Next AI Platform DeepSeek Harness and OpenAI's Codex harness are now open source. But the larger story is not simply that two more repositories became public. Zhihu contributor 第欧根尼 argues that Agent products are beginning to unbundle. The future may be less about choosing one monolithic Agent app and more about assembling a runtime, model router, scenario-specific distribution, and trusted plugin stack. 1️⃣ A harness is becoming part of the model The author's first hypothesis is that a harness will no longer be just a frontend that exposes model capabilities. It will become a framework that co-evolves with the model. The same model may perform very differently inside its official harness than inside a third-party implementation. Context management, tool descriptions, task decomposition, verification, and retry policies all influence the model's behavior. This leads to an important conclusion: A model and its Agent loop can produce better results together than the model can deliver on its own. The competitive unit is therefore shifting from the model alone to the model-harness system. 2️⃣ Agents are becoming lighter and more distributed The author's second observation comes from the evolution of MCP, Cloudflare's Agent infrastructure, and the growing demand for programmable workflows. He expects Agents to become: 🔹 Smaller and more specialized 🔹 Easier to customize through code 🔹 More independent from monolithic apps 🔹 Numerous enough to run as lightweight background workers Current products such as Kimi Work or WorkBuddy still control much of their unique behavior internally. Users cannot easily modify them or embed their complete workflows inside an enterprise system. But market demand is moving toward more flexible forms: plugins lighter than standalone apps, Code Mode more powerful than static skills, and large numbers of low-overhead Agents running simultaneously. That helps explain why vendors are opening their harnesses now. 3️⃣ The harness becomes a microkernel DeepSeek Harness treats the harness as something closer to a microkernel plus a distribution. The base runtime becomes thinner. It retains only the functions every Agent needs: 🔹 Plugin loading and lifecycle management 🔹 Event routing 🔹 Permissions and state 🔹 Execution protocols 🔹 Session and context infrastructure Research, coding, office work, and customer service are then assembled through different plugin bundles. The author sees OpenAI's Codex harness moving in a broadly similar direction, even if it uses different terminology. In this model, users may stop choosing a single Agent product. Instead, they choose: runtime + model routing + scenario distribution + organization plugins DeepSeek Harness and Codex become open runtimes on which many different Agent products can be built. 4️⃣ Five changes follow from this architecture 🔹 Plugin count stops being meaningful Prompts, skills, MCP services, and harness plugins can multiply quickly. The difficult problem will not be finding more plugins, but deciding which ones are trustworthy. Security review, provenance, compatibility, maintenance, and permission control become the real barriers. 🔹 Models become replaceable execution resources If context and data remain inside the harness, the runtime can route different tasks to different models. A strong model may handle planning and review, while cheaper models perform repetitive execution. Switching models becomes a runtime decision rather than a full migration. 🔹 The Agent Loop becomes the main optimization target As model capabilities converge, user experience may depend more on the surrounding loop: When should context be compressed? When should a task be split? What should be remembered? How should results be verified? Improving these decisions may create more value than replacing the underlying model. Models trained to cooperate with a particular harness could gain a significant advantage. 🔹 Skills, MCP, and plugins form a compatibility layer The market is unlikely to accept a different extension format for every platform forever. Competition will shift from “does this platform support plugins?” to “how many ecosystems can it support without degrading the experience?” 🔹 Personal runtimes separate from enterprise control planes Individuals need flexible local Agents. Enterprises need governance, private marketplaces, permission policies, observability, and integration management. These will become distinct product layers, even when they share the same open runtime. 5️⃣ Existing Agent products will defend through ecosystems The author expects products such as WorkBuddy to expose compatibility layers without fully opening their core runtime. They may quickly announce support for DeepSeek Harness plugins, Agent Skills, and more MCP services. But these capabilities would likely enter through adapters rather than replace the underlying harness. They may also build private enterprise plugin marketplaces. The more open the ecosystem becomes, the more companies need vendors that can absorb integration and security risks. Distribution remains another moat. WorkBuddy can connect deeply with WeChat, WeCom, and Tencent Docs. An open harness may reproduce its plugins, but it cannot quickly reproduce users' work relationships and established business entry points. Alibaba has a different advantage. The author expects it to use Alibaba Cloud and Bailian to provide a managed harness control plane, turning cloud infrastructure into the runtime layer for enterprise Agents. 6️⃣ This looks like the Android/AOSP moment for Agents The current market resembles the early Android ecosystem. An open foundation can stop hundreds of teams from rebuilding the same runtime. But publishing reference code is not enough to create an Android-scale platform. The next six months may decide whether these projects converge into a durable ecosystem. They need stable interfaces, trusted plugin infrastructure, and vertical software teams willing to maintain real products on top of the open runtimes. The decisive question is not whether DeepSeek or OpenAI has released the better harness today. It is whether the industry can turn open harnesses into a shared Agent platform, rather than another collection of incompatible reference implementations. 🔗 Full analysis: #DeepSeekHarness# #OpenAI# #Codex# #AIAgents# #AgentInfrastructure# #MCP# #OpenSourceAI#

If your agent executes code, you want to run it in an isolated environment. I learned this the hard way. I also know a couple of people who let OpenClaw go nuts and had to wipe out their laptops after a few days to clean up the mess. Two rules you must always follow: 1. Agent that runs code → isolated environment 2. Agent that controls a browser → isolated environment But that generates a new problem: that environment has to start quickly, use as little memory as possible, and give you enough control to operate it. This is hard to achieve, but you don't need to reinvent the wheel: Cube Sandbox is an open-source MicroVM sandbox developed by Tencent Cloud. Think of it as a production-grade agent infrastructure that you can use to run agents at scale. Cube Sandbox uses RustVMM and KVM to run each sandbox with hardware-level isolation. • It cold-starts in under 60ms • It uses less than 5MB of RAM overhead • You can launch tens of thousands of sandboxes within a minute Here is what you can do with it: 1. Capture, clone, and restore sandbox state with snapshots and rollback 2. Deploy sandbox clusters through Kubernetes 3. Run workloads on ARM infrastructure 4. Observe what happens inside agent executions 5. Govern ingress and egress traffic 6. Persistent volumes and cross-machine pause/resume Here is the GitHub repository: You can inspect the code and try it out with your own agent workflow. #ad#

ChainBase's (@ChainbaseHQ) AI agent gateway AgentKey (@agentkeyapp) has integrated RootData's data capabilities into its service marketplace. Users can now access structured data on more than 20,000 crypto projects through the platform, including project overviews, team backgrounds, funding history, key event calendars, and the latest news. The integration provides AI agents with richer, real-time off-chain research data, enabling more efficient access to multi-dimensional information for automated strategies and investment research. It also deepens AgentKey's role as AI agent infrastructure. For RootData, the collaboration further expands our presence in AI scenarios and strengthens our position as a trusted data infrastructure provider for the AI agent ecosystem.

"𝟗𝟎% 𝐀𝐈-𝐰𝐫𝐢𝐭𝐭𝐞𝐧" ≠ 𝐚𝐠𝐞𝐧𝐭𝐢𝐜 𝐞𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠 Without autonomous-agent infrastructure (multiple parallel agents per engineer) you're still capped by human attention. That's an IDE plugin with a good marketing line. AI-coding tools alone are not what transforms developer teams—it’s about the operating model along with it. Infrastructure enables the speed, but the people and org structure decide whether it actually compounds. Learn more with 𝐓𝐡𝐞 𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐀𝐰𝐚𝐤𝐞𝐧𝐢𝐧𝐠 →

Why @Box chose Deep Agents: 1️⃣ Complete model agnosticism Customers can choose LLM providers, and Deep Agents allows this flexibility at the platform level. 2️⃣ Speed of iteration Open agent harness = less time spent on building core agent infrastructure, more time solving enterprise-specific problems. Inside the Box Agent ⤵️

What is more fulfilling than a month full of building and shipping? From new integrations and production deployments to agent infrastructure and expansion to Intelligence Oracle, Cournot has continued turning its roadmap into working systems. Key updates 👇: 🔸 Cournot's AI-Native Oracle has been serving as Official Resolution Partner for @42Space. Cournot supports the settlement process with real-time monitoring, evidence-backed evaluation, and transparent resolution based on predefined rules. 🔸 Cournot is now embedded in @MyriadMarkets’ core resolution workflow as an independent AI voter, delivering fast, neutral, evidence-backed judgments. 🔸 Launched the Intelligence Oracle during the 2026 World Cup: • 3,100+ live markets affected • ~$498.4M notional influenced • 95%+ of registered wallets reached the leaderboard 🔸Cournot's Resolution Oracle Key Metrics: • $1.39B notional influenced • $9.23B shadow notional influenced • 5,625 resolutions • 99.32% accuracy • <8 min latency 🔸 Cournot has launched an AI-native oracle agent on @virtuals_io as an independent evaluator for real-world outcomes. Aligned with the ERC-8183 evaluator standard, the agent reviews evidence, interprets task conditions, and produces transparent, verifiable determinations for agent-to-agent commerce. 🔸 Cournot has built an AI-powered voter agent on @BNBCHAIN using the BNB Agent SDK. Designed for agent-to-agent commerce, it acts as an automated dispute evaluator, reviewing submitted work, checking it against agreed requirements, reaching an AI-based judgment, and casting the result onchain without a human moderator. 🔸 Updated Whitepaper: From redefining our thesis Oracle Evolution to expanding from Resolution Oracle into a broader verifiable intelligence stack, Cournot's AI Native oracle is built to serve multiple TAM: prediction markets, RWA & collectibles (price feeds, VRF, etc.), trading intelligence & signals, DeFi and beyond. Upward and onward.

Ask an AI to write a paper and it will happily report favorable numbers for experiments it never ran — this "plausible fabrication" has been the real wall for research automation. Earlier automated-research systems poured effort into standing up dedicated agent infrastructure and forcing a single clean story to completion. But somewhere in the long generation, factual integrity would fray, and unsupported claims slipped straight into the paper. Behind the flashy results, the foundation of reliability was quietly missing. Spark-to-Paper flips the framing. Instead of a new agent platform, it implements paper generation as 13 composable skills running inside Claude Code. The key is separation: context-dependent judgment goes to the model, while citation checks and LaTeX compilation go to deterministic code. Experiments fix their metrics and table structure before results are observed, blocking post-hoc adaptation. And a "Self-Refutation Loop" — endlessly revising a hypothesis the evidence keeps rejecting — is capped at seven cycles, recording failure as failure and moving on to a different direction. The results speak clearly. Citation validity hits 99.5%, above the human-preprint average, and fabrication detection climbs from 14% single-pass to 92% with the full adversarial-review stack. At roughly $8.1 and 3.2 hours per manuscript, you get a paper complete with editable figures. Rather than dressing every attempt up as success, it centers evidence integrity and the acceptance that "not every research direction works." A thought-provoking blueprint for reliability in long-horizon AI. Spark-to-Paper: End-to-End Research Paper Generation as a Composable Skill #AIResearch# #AgentSkills#