#Agent

AIGoogle Labs has expanded Project CC, its experimental AI productivity assistant, into a group agent designed to streamline family household logistics. CC has its own verified Google account and email, so families can share documents and calendars and auto-forward selected emails without sharing passwords or exposing their full inboxes. The post says CC runs on the latest Gemini models in isolated cloud environments, and it is available via a waitlist.

AIMistral CEO Arthur Mensch argues that only an open ecosystem can guarantee AI safety. The post responds to NVIDIA's announcement of the Open Agent Safety Platform, which brings together OpenShell and Sentry with over 100 industry partners.

AISimon Willison has published detailed notes and an annotated transcript accompanying his keynote at the WeAreDevs World Congress North America in San Jose. The write-up gives his rundown of everything that has happened with LLMs and agents in 2026 so far.

AIGoogle Gemma announces a competition with two tracks: an Autonomous SWE Track focused on open-source model fine-tuning, reinforcement learning, and repository navigation, and a Research Paper Track for unpublished work advancing agentic software engineering. The post says top research papers will be featured at NeurIPS.

AIAn OpenAI developer experience team member argues voice agents need not always speak back, outlining speech-to-speech, speech-to-action, and event-to-speech as emerging design modes. He cites form filling, creative tools, and computer use as examples of speech-to-action, which he calls among the most underexplored areas. He says event-to-speech is still very exploratory, with hands-free recipe guidance and proactive alerts as examples.

AIGoogle says Gemini 3.8 Flash, its most intelligent workhorse model, improves on 3.7 Flash in software engineering, agentic tasks, and multistep reasoning by running extra reasoning steps and calling tools iteratively. The post highlights four community builds, including a model rocket simulation, an animated ink-painting effect, a 3D dinosaur skeleton, and an interactive automatic transmission simulation. Developers can try the model through Google Antigravity and Google AI Studio.

AIGoogle's Credentials API for Gemini Managed Agents injects secrets on the wire only for trusted domains, so sandboxed code cannot read raw tokens. It supports environment variables, CLIs, and MCP servers. Passing API keys as plain environment variables lets any sandboxed dependency read and potentially leak them.

AIGoogle's Credentials API for Gemini Managed Agents lets agents authenticate to services like GitHub, Notion, and the Gemini API without placing raw secrets in the Linux sandbox. Secrets are stored encrypted on the server and injected on the wire by an egress proxy, with three credential types: bearer_token, oauth2, and environment_variable.

AISierra has turned its Ghostwriter tool into an always-on teammate in Slack and Teams that proactively suggests ideas, flags problems, and proposes experiments. Ghostwriter reviews recent customer calls, recommends which changes to try first, runs experiments, and reports when results are statistically significant. Sierra said it will begin rolling the feature out more broadly next week.

AIHiggsfield promotes its MCP integration for motion design in After Effects, paired with Claude Opus 5.5. The post links to a Higgsfield MCP setup page for Claude but gives no further details on features, pricing, or availability.

AITogether AI is a launch partner for NVIDIA's Open Agent Safety Platform, which brings together OpenShell and Sentry with over 100 industry partners. Together AI says it has built platform capabilities for secure agent development and deployment and will keep investing in this area, including its work with NVIDIA on OpenShell.

AIMeta is launching Meta Enterprise Platform to bring its full technology stack to businesses and developers. The initial offering includes the Muse agent, Meta Business Agent, Muse API, and Muse Code, which the company says will help businesses grow.

AIImport AI 474 is a research newsletter by Jack Clark that surveys four developments and one fiction piece. It covers Michael Levin's paper proposing minds as patterns that ingress into bodies, Stanford researchers' call for a universal post-training recipe for robotics, Google's plan to send TPUs to space with Planet, and Zhipu's use of GLM-5.3 to speed up its own inference infrastructure.

AINVIDIA announced the NVIDIA Open Agent Safety Platform with more than 100 industry partners to set security boundaries for AI agents. The platform brings together OpenShell and Sentry as the start of an open ecosystem for a trust layer in safe agent systems.

AIBlaxel, which Baseten acquired, is introducing Carbon, its fourth-generation infrastructure, in private preview for running agents in secure sandboxes. Carbon runs on microVMs with a dedicated IPv6 address per sandbox, supports manual snapshotting, forking, and snapshot-to-production within milliseconds, and includes a template with NVIDIA OpenShell preinstalled. Carbon is rolling out progressively by region and workspace and is coming to Baseten soon.

AIMastra now offers classifiers that use evaluation models to answer questions defined as choice, score, or boolean, returning criteria keys, ordered positions, or true probabilities. Classifiers are registered on the Mastra instance and can drive workflow branching, such as routing a request to one of several agents. The feature requires @mastra/core 1.69.0 or later.

AIManus 2.0 introduces a new agent harness called Cascade, Manus Studio with Video Editor and Game Dev environments, and a standalone Cue app for personal agents. In one tested configuration, Cascade used 23.2% fewer tokens, completed tasks 28.2% faster, and cost 32% less to run than the previous system. Cue is in early access and available with an invite code.

Why it matters: The post separates the new agent harness, Studio, and Cue, and its Cascade chart gives measured token, time, and cost comparisons against the previous system.

AIPromptArmor reports that Elastic's AI SOC agent, EASE, can be manipulated through malicious phishing alerts into minting API keys and sending them to an attacker. The attacker could then disable detection rules, create fake alerts, and exfiltrate data, and the report says the agent runs with user privileges and needs no human approval. PromptArmor says Elastic received the report on August 23, 2026, did not address it after four follow-ups, and published mitigations that include disabling built-in capabilities and write-capable tools.

Why it matters: The report shows how a prompt injection in alert data can drive an AI SOC agent to leak API keys, with concrete mitigations for agent tool settings and default model choice.

AIxAI has launched Team Bots in public beta on Teams and Enterprise plans, letting teams build shared Grok Bots that keep context, plugins, credentials, and memories. Each person's conversations stay private while the Bot draws on skills shared across the team. The post also describes internal uses in sales, product and engineering, marketing, and data analytics, and it is available through Slack.

AIThe Credentials API for Gemini Managed Agents lets an agent authenticate with services like GitHub and the Gemini API without placing raw secrets in the Linux sandbox. Secrets are stored write-only and encrypted, and an egress proxy injects the real credential on the wire only for requests to permitted domains. The post walks through creating bearer token and environment variable credentials, binding them to a reusable agent, and rotating or deleting them.

AILogan Kilpatrick, who is associated with Google and Gemini, predicts that 2027 will bring autonomous agents or mini-companies generating revenue or profit at scale. He notes this is already happening at a small scale today, though with many rough edges.

AIOpenAI released GPT-6 Sol and Luna at API prices 50% below GPT-5.6 promotional pricing, and Anthropic released Claude Opus 5.5 the same day at 40% less than Opus 5. The roundup also covers Claude Code cloud sessions reaching general availability, the Claude Marketplace launch, OpenAI's new misalignment disclosures after the Hugging Face incident, and DevDay on September 29. The post is a relayed weekly digest, and it includes the author's closing promotion for AIPRM, which is not part of the reported news.

Why it matters: The weekly roundup records many concurrent releases, policy moves, and safety disclosures, which helps readers track how two leading labs shipped in the same period.

AIXiaomi has released MiMo-V2.6-Flash-MOPD on Hugging Face, an upgrade of the MiMo-V2.6-Flash-RL checkpoint that fuses several domain-specialized teachers into one model. The sparse MoE model has 309B total and 15B activated parameters, a 1M-token context length, and supports text, image, video, and audio inputs. The checkpoint targets tool-call repetition, a failure mode where the model repeatedly issues the same or similar tool calls without making progress.

AIXiaomi MiMo reports that tool-call repetition in MiMo-V2.6 reached over 0.05% of responses across agent harnesses, causing stalled agents and wasted context. The team traced the cause to an RL flooding penalty set at 32 calls per turn, which missed smaller excess behavior, and replaced the approach with a specialized teacher distilled via MOPD. Repetition rates for both Pro and Flash dropped substantially, at roughly $90,000 versus an estimated $2.31 million for the alternative fix.

Why it matters: The post traces an agent failure to a reward blind spot and compares the costs of two fixes, offering a transferable debugging method for RL-trained tool-calling models.