Office24_ProPlus2024VL_MAK_AE2 Description: Office24_ProPlus2024VL_MAK_AE2 SubType: X23-57852 LicenseType: Volume:MAK MAKCount: 1228 Time: 30/07/2026 12:58:50 (GMT+7) Key: 6VKWD-CNRG3-MJHBQ-9MRHQ-H26J3

We've investigated a few messages about codex usage limits being different. That's not something we change without engaging the community and being transparent. What we did see is that when talking to affected users many were using sub2api. Converting a subscription into api traffic to then re-serve or share across many users is not something we support and this type of usage gets flagged by our fraud-prevention systems. You are completely fine if you use your subscription through Sign in With ChatGPT, either through the official clients or through one of the many OSS clients (Pi, OpenCode, ...) that support signing in with your account and using your included usage.

# Practical ways to use the Claude Agent SDK 🤖 Delegate specialized tasks to sub-agents and keep your main context lean. Sub-agents delegate specific subtasks to specialized agents, enabling parallel and expert processing without polluting the main agent's context. 📌 Title: Sub-agents in the SDK 🔗 URL: 🧩 Overview Define specialized agents via the `agents` parameter. The main agent calls them through the `Agent` tool. Each sub-agent gets its own prompt, tool restrictions, and model settings. Results return as summaries to the main agent. 🛠 How to use it Define sub-agents in the `agents` parameter using `AgentDefinition` with `description`, `prompt`, `tools`, and `model`. Include `"Agent"` in `allowed_tools` to auto-approve sub-agent invocations. Each sub-agent gets its own tool restrictions, prompt, and model settings. 🏗 Practical usage - Delegate large file exploration to a `research-assistant` sub-agent, returning only summaries to the parent. Keeps main context lean. - Run `style-checker` / `security-scanner` / `test-coverage` concurrently, reducing code review time from minutes to seconds. - Give a `database-migration` sub-agent SQL best practices and rollback strategies in its prompt for expert handling. - Use factory functions to dynamically select `model: "opus"` vs `sonnet` based on runtime conditions. 💡 Use cases 🔍 Delegating large-scale file exploration ⚡ Parallel code review (style/security/coverage) 🎯 Expert task splitting across specialized agents ⚠️ Watch out Sub-agents cannot spawn their own sub-agents (one level only). For large-scale orchestration, use the `Workflow` tool (TS v0.3.149+). Capture `agentId` and `session_id` for follow-up via `resume`. #ClaudeAgentSDK# #AI#

# Practical ways to use the Claude Agent SDK 🤖 Delegate specialized tasks to sub-agents and keep your main context lean. Sub-agents delegate specific subtasks to specialized agents, enabling parallel and expert processing without polluting the main agent's context. 📌 Title: Sub-agents in the SDK 🔗 URL: 🧩 Overview Define specialized agents via the `agents` parameter. The main agent calls them through the `Agent` tool. Each sub-agent gets its own prompt, tool restrictions, and model settings. Results return as summaries to the main agent. 🛠 How to use it Define sub-agents in the `agents` parameter using `AgentDefinition` with `description`, `prompt`, `tools`, and `model`. Include `"Agent"` in `allowed_tools` to auto-approve sub-agent invocations. Each sub-agent gets its own tool restrictions, prompt, and model settings. 🏗 Practical usage - Delegate large file exploration to a `research-assistant` sub-agent, returning only summaries to the parent. Keeps main context lean. - Run `style-checker` / `security-scanner` / `test-coverage` concurrently, reducing code review time from minutes to seconds. - Give a `database-migration` sub-agent SQL best practices and rollback strategies in its prompt for expert handling. - Use factory functions to dynamically select `model: "opus"` vs `sonnet` based on runtime conditions. 💡 Use cases 🔍 Delegating large-scale file exploration ⚡ Parallel code review (style/security/coverage) 🎯 Expert task splitting across specialized agents ⚠️ Watch out Sub-agents cannot spawn their own sub-agents (one level only). For large-scale orchestration, use the `Workflow` tool (TS v0.3.149+). Capture `agentId` and `session_id` for follow-up via `resume`. #ClaudeAgentSDK# #AI#

@qrazhan_ultd @komeng_ho @lowbrowkid @ChinemeRay52296 @fuadm2550 @tombanner @Jogvrel @thedrillas @LVIronside @Herica97440089 @Tps_85 @SwaggyZhoo @BruceWillisX @janusicay @DianaMendo26779 @abdullah_m28525 @1c0fd7eee4f2437 @1000xhustl62000 @I_am_Subcaptain @MercishoWillfre @ysfgnydn61 @zhang_zong86842 @knight_kevin @markwa14 @w3roman @ModelskiBartosz @cancinc @JohnPar22010984 @teampsgroup @dancoombs45 @MokaramShahram @shockxbt @MDDatta153324 @PANDAO11032252 @Cam48C @trigga_AD @theBTCMD @nep2un3 @matty_xyz @Dexterumit @ayesha4160 @scubbyo @dickySOL @sacha06400 @Mike_Moons89 @userloli00 @Lt_QRF @0xXian @traceyarth @mikeycrypto19

Hermes can now break a big job into phases, then manage the individual tasks underneath each one. Its todo system is no longer limited to one long flat checklist. So instead of: ☐ Build homepage ☐ Add authentication Hermes can now structure the same work like this: ► Build homepage ↳ Create hero section ↳ Add navigation ↳ Build pricing section ↳ Add contact form ► Add authentication ↳ Create sign-in page ↳ Add account creation ↳ Connect auth provider ↳ Handle password reset That is the actual change. A task can now become a parent with its own nested subtasks underneath it. So when Hermes is working through something larger, it no longer has to flatten every piece of work into one list. It can keep the major phases clear, break each one into the work required underneath it, and manage those subtasks individually as the job progresses. And this is more than visual organization. Hermes keeps the parent/subtask relationships intact as work gets updated. If context is compressed during a longer session, that hierarchy is preserved too, so unfinished subtasks still retain the larger task they belong to. Even a completed parent can remain visible while work underneath it is still active. That makes the todo system much more useful for larger coding jobs, research, content production, automations, migrations, business workflows, and anything else where one task really contains several smaller ones.

🤖 What changes when a robot doesn't just act, but thinks ahead and predicts the future before it moves? Here's an embodied foundation model trained on over 37,000 hours of data. Title: GigaBrain-0.7: Scaling Embodied Foundation Models to Emergent Capabilities with a Three-System Architecture URL: GigaBrain-0.7 proposes a three-system architecture that cleanly separates yet integrates understanding, prediction, and action. Three highlights stand out. 🧠 System 2: Understanding and planning Built on PaliGemma2 (3B), it interprets the current visual scene and long-horizon instructions, using chain-of-thought reasoning to break "hang it on a hanger" into a subtask like "first pick up the red hat." 🔮 System 3: Prediction and evaluation A GigaWorld-1-based model (5B) predicts future visual states as video, extracting the final frame as a "subgoal image." It also scores task progress as a value signal that guides the action system. 🦾 System 1: Action and control A Mixture-of-Transformers setup generates continuous action chunks, using a unified embodiment representation across 16 robot types for cross-robot generalization. What stands out most is that larger-scale pretraining consistently lowered validation loss, with signs of genuinely emergent embodied capabilities showing up along the way. #Robotics# #EmbodiedAI#

TL;DR: Three failure modes of long-horizon agents — compounding errors, context rot, and task-state loss — solved structurally through a Manage-Execute-Audit (MEA) loop. WeaveBench PassRate: 51.8% → 80.7%. LongHorizon-Harness: Advancing Long-Horizon Agents for Real-World Tasks Key points: 🔧 Manager: Maintains explicit task state outside the execution trajectory. Constructs subtask contracts specifying goals, acceptance criteria, constraints, and evidence. ⚡ Executor: Runs each subtask in a fresh, budget-bounded context — no prior trajectory history passed in. 🔍 Auditor: Post-execution read-only inspection, independent of the executor. Reports completion, integrity, and state updates; serves as persistent cross-round memory. 🖥️ GUI/CLI hybrid: Manager routes tasks to the right interface; AgentAdapter swaps in Claude Code, Codex CLI, Hermes Agent without modifying native loops. 📊 WeaveBench: PassRate 51.8% → 80.7%; Design +60pp, Spatial/3D +50pp. 🤖 OSWorld 2.0: Qwen 3.7-Plus 2.8% → 8.3% (3×); Claude Opus 4.7 20.6% → 35.3%. 💻 Terminal-Bench: 69.7% → 77.2%, with 24% fewer tokens consumed. 💡 Manager overhead: only 2–8% of total tokens. Auditor carries 19–38% — the primary cost of reliability. "Agent capability is a property of the complete model–harness system, not the model alone" — that framing changes how you should think about long-horizon agent design. #AIAgents# #LLM#