GBrain 0.46.28: 111 verified issues and remote private-page lockdown
GBrain 0.46.28: private-page remote lockdown, daemon env file superseding interim key channels, probe-liveness/queue-unlatch, and a model-tier vs models.default precedence break.
News
The Agent Harness Becomes an Attention Interface
Latent Space essay: as models absorb harness logic, the leftover product is a policy surface for scarce human attention.
Hermes Agent 0.20.5 tags a 323-PR window; Ox Alpha lands via OpenCode
Nous tags Hermes Agent 0.20.5 as the 323-PR snapshot since 0.20.4; Teknium says Ox Alpha now runs in Hermes via OpenCode and OpenRouter.
Claude Security scans now run on Mythos 5 for Enterprise
Anthropic puts Mythos 5 behind Claude Security scans for Enterprise: CWE findings, human-reviewed patches in Claude Code on the web, plus a $35M OSS fund.
monday.com's Claude rebuild: five lessons after 5M agent interactions
Anthropic publishes monday.com's agent-first rebuild: four Claude integration paths and five lessons after a claimed 5 million agent interactions.
OpenAI: Hosts Own the Interface, Codex Owns the Agent Loop
OpenAI's Codex-as-a-platform post tells builders to embed the existing Codex harness in their own UIs, with exec, SDK, and app-server as the three integration layers.
Command Code 1.31.0 ships free Ox Alpha; Desktop 0.1.14 adds GitHub skill installs
Command Code CLI 1.31.0 adds free Ox Alpha, a 1M-context model on every plan. Desktop 0.1.14 installs skills from GitHub with confirmation.
Pydantic AI Harness 0.24.0 adds PlaywrightBrowser
Pydantic AI Harness 0.24.0 ships PlaywrightBrowser via PR #420: a stateful Chromium capability with allowlists, screenshots, and JS—in the harness, not pydantic-ai core.
Pydantic AI 2.33.0 requires Anthropic SDK 1.0 / httpx2
Pydantic AI 2.33.0 requires anthropic>=1.0.0 and httpx2 after Anthropic 1.0 broke unpinned installs; pin anthropic<1 to stay on older pydantic-ai.
NVIDIA NemoClaw 0.0.113 tightens sandboxes, adds experimental Hermes Google Chat
NemoClaw v0.0.113 (Aug 20) improves sandbox recovery, adds experimental Hermes Google Chat with keys outside the sandbox, and changes the Gemini default.
Slack Code opens dedicated channels for teams and coding agents
Slack Code spins tagged coding-agent work into dedicated, archivable channels. Claude, Devin, Copilot, and Vercel are live; ChatGPT timing is inconsistent across Slack and Salesforce.
Codex CLI 0.149.0 adds an agents dashboard and codex queue
OpenAI Codex CLI 0.149.0 ships a TUI agents dashboard, codex queue for local/remote sessions, cwd commands, and broader doctor diagnostics.
Claude Code 2.1.238: memory-growth fix and self-hosted runner controls
Claude Code 2.1.238 stops unbounded session growth by dropping off-screen subagent tool results, and adds runner SIGTERM deferral plus plugin header helpers.
OpenClaw demos unreleased Control UI, shared gateway, Mac onboarding
OpenClaw’s ClawCast Episode 8 demos an unreleased Control UI, shared RoboClaw gateway, and Mac onboarding; maintainers say the next stable is waiting on stability.
Vercel Agent public beta in Slack: incidents, PRs, rollback
Vercel Agent is in public Slack beta for Pro/Enterprise: diagnose production issues, open PRs, and roll back after human approval of a plan.
Mastra MCPServer can opt into MCP 2026-07-28 and multi-round elicitation
Mastra MCPServer can set protocolVersion 2026-07-28 for stateless MCP, dual-era HTTP, and multi-round elicitation. Core is 1.60.0 on npm.
Nous: Hermes Cloud idle instances billed at 3 cents per day
Hermes Cloud documents $0.03/day idle storage and $0.29–$1.09/day running compute; Nous says the cheapest plan is 3¢ idle, 29¢ nonstop.
Codex CLI 0.148.0 adds TUI Markdown /export and session fork
Codex CLI 0.148.0 ships Markdown /export, session fork plus TUI archive/restore.
VS Code 1.134: chat grids, prompt timeline, Find in chat
VS Code 1.134 (Aug 19) adds side-by-side agent chats, a prompt timeline, and Find in chat.
Claude Code 2.1.237 adds Concise output style and gateway cache fix
Claude Code 2.1.237 adds a Concise output style that leads with results, plus a prompt-cache fix for LLM gateways and custom base URLs.
Claude Managed Agents: self-hosted memory, web domain controls, Console rebuild
Anthropic lets self-hosted Managed Agents attach memory, restrict web_search/web_fetch domains, and inspect multi-agent sessions in a rebuilt Console.
NVIDIA open-sources SkillEvaluator; Hermes scans skill installs before confirm
NVIDIA open-sources SkillEvaluator after 300+ skill benchmarks; Hermes now advisory-scans hub installs for PII, secrets, and smuggling.
Claude Code 2.1.236 adds ANTHROPIC_DEFAULT_MODEL and idle cross-session notify
Claude Code 2.1.236 adds ANTHROPIC_DEFAULT_MODEL, one-shot idle notify, tighter macOS sandbox denies, and auto-mode classifier improvements.
Grok 4.6 is generally available on Amazon Bedrock
Grok 4.6 is GA on Amazon Bedrock: 500k context, xhigh reasoning, $2/$6 per 1M tokens, plus US and global cross-Region inference.
Anthropic graduates Agent Skills API and Files API to GA
Anthropic’s Aug 19 notes make Agent Skills, the Files API, and Enterprise Admin user-management generally available without beta headers.