{"date":"2026-10-26","title":"What happened in AI — Oct 26, 2026","generated_at":"2026-10-26T22:17:17Z","intro":["Agent security led the day: a malicious AGENTS.md planted in a Downloads folder hijacked coding agents, with an npm install attack reproduced against OpenAI Codex.","Builders also shipped harnesses, memory layers and a Git forge for agents, while OpenAI released GPT-6.1 Terra for computer use."],"highlights":["Planted AGENTS.md in Downloads steered Codex into installing a rogue npm package.","Google Cloud adds IAM helper agent skills and MCP tools for scoped agent identities.","Entire launches a Git forge built for agent-human collaboration.","Databricks AI Search now covers Streaming Tables, Materialized Views and Iceberg v3+ tables.","OpenAI introduces GPT-6.1 Terra for fast computer use."],"article_count":23,"categories":[{"name":"Coding agents inherit the attack surface of their inputs","slug":"coding-agent-attack-surface","summary":"Two reports show a planted instruction file in Downloads steering coding agents, while IAM and IaC tooling targets the same access risk.","articles":[{"title":"AI coding agents hijacked using AGENTS.md file in Downloads folder","summary":"Malicious AGENTS.md in a Downloads folder tricked an agent into installing a rogue npm package; reproduced against OpenAI Codex, with a related Claude Code OAuth-token leak.","source":"hackernews_ai","url":"https://www.bleepingcomputer.com/news/security/ai-coding-agents-hijacked-using-agentsmd-file-in-downloads-folder/","published":"Mon, 26 Oct 2026 14:01:56 +0000"},{"title":"Markdown Is the New Executable: Attackers Drop Agent Instructions into Downloads","summary":"Forever caught live malware planting an agent-instruction file in Downloads, evidence that markdown files now work as an execution vector.","source":"hackernews_ai","url":"https://forever.security/blog/hijacking-ai-agents-via-downloads-folder/","published":"Mon, 26 Oct 2026 14:04:26 +0000"},{"title":"Scale agents securely with IAM Helper Agent Skills and MCP Tools","summary":"Google Cloud pairs IAM helper agent skills with MCP tools so agents provisioning cloud resources and writing Terraform run under scoped identities.","source":"google_cloud_blog","url":"https://cloud.google.com/blog/products/identity-security/scale-agents-securely-with-iam-helper-agent-skills-and-mcp-tools/","published":"Mon, 26 Oct 2026 16:00:00 +0000"},{"title":"Using AI Verification and Fix Loop for IaC (Terraform, OpenTofu, Pulumi)","summary":"Masterpoint argues IaC code generation is now cheap while review is the bottleneck, and proposes a verify-and-fix loop for Terraform, OpenTofu and Pulumi.","source":"hackernews_ai","url":"https://masterpoint.io/blog/ai-verification-fix-loop-iac/","published":"Mon, 26 Oct 2026 17:34:31 +0000"}]},{"name":"Agent harnesses, memory and team tooling","slug":"agent-harness-memory-tooling","summary":"Builders keep shipping harnesses, memory layers and review surfaces around coding agents instead of waiting on models.","articles":[{"title":"Article: The Evolution of Agentic Harnesses: From Chains to Graph to Code","summary":"InfoQ traces agentic harness design from linear chains to graphs to code-driven control flow.","source":"infoq_ai_ml","url":"https://www.infoq.com/articles/agentic-ai-architecture-harness-evolution/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=AI%2C+ML+%26+Data+Engineering","published":"Mon, 26 Oct 2026 11:00:00 GMT"},{"title":"Humanlayer: Way to get AI coding agents in complex codebases","summary":"HumanLayer runs Claude Code, Codex and OpenCode on your machines and syncs tasks, plans and reviews across desktop, web and a daemon.","source":"hackernews_ai","url":"https://github.com/humanlayer/humanlayer","published":"Mon, 26 Oct 2026 16:32:14 +0000"},{"title":"The Git Forge for Agent-Human Collaboration","summary":"Entire launches a Git forge built for agent-human collaboration, with an open ecosystem of partner tools.","source":"hackernews_ai","url":"https://entire.com/blog/the-git-forge-for-agent-human-collaboration","published":"Mon, 26 Oct 2026 17:23:11 +0000"},{"title":"Show HN: Graph-MIND – Local memory for AI coding agents, no LLM calls to save","summary":"Local agent memory that costs zero model tokens to save; claims 88.8% on LongMemEval, beating Mem0 on the same tests (self-reported).","source":"hackernews_ai","url":"https://github.com/goyohan0611-png/graph-mind","published":"Mon, 26 Oct 2026 12:30:12 +0000"},{"title":"Show HN: Mainplane – Agent in Go that lives on your old PC, drives your laptop","summary":"Go agent harness from a former Browser Use engineer that runs on an old PC and drives a laptop.","source":"hackernews_ai","url":"https://github.com/mainplane-ai/mainplane","published":"Mon, 26 Oct 2026 16:37:03 +0000"}]},{"name":"Inference scheduling and serving tests","slug":"inference-serving-testing","summary":"Teams are building inference primitives and per-change GPU tests for self-hosted coding models.","articles":[{"title":"Show HN: Resin, fast native scheduler for LLM inference engines","summary":"Rust scheduler, KV cache and metrics server as building blocks for LLM inference; the author notes it is not production-quality.","source":"hackernews_ai","url":"https://github.com/ekzhang/resin","published":"Mon, 26 Oct 2026 17:06:14 +0000"},{"title":"How we test LLM serving changes for coding agents, two hours of GPUs per change","summary":"Upbound describes testing each LLM serving change for coding agents with two hours of GPUs per change.","source":"hackernews_ai","url":"https://www.upbound.io/blog/drinking-our-own-champagne-part-5-what-does-it-take-to-grow-a-self-hosted-coding-model-from-pilot-to-team","published":"Mon, 26 Oct 2026 15:12:56 +0000"}]},{"name":"Data and model releases","slug":"data-model-releases","summary":"Databricks extends AI Search to live data, and OpenAI ships a new computer-use model.","articles":[{"title":"Building live intelligence: Databricks AI Search supports Streaming Tables, Materialized Views, and Managed tables using Iceberg","summary":"AI Search now indexes Streaming Tables, Materialized Views and Iceberg v3+ managed tables alongside Delta, keeping live state searchable.","source":"databricks_blog","url":"https://www.databricks.com/blog/building-live-intelligence-databricks-ai-search-supports-streaming-tables-materialized-views","published":"Mon, 26 Oct 2026 18:03:59 GMT"},{"title":"Introducing GPT-6.1 Terra","summary":"OpenAI introduces GPT-6.1 Terra, positioned as fast, capable computer use for everyone.","source":"openai_blog","url":"https://openai.com/index/introducing-gpt-6-1-terra","published":"Mon, 26 Oct 2026 00:00:00 GMT"},{"title":"How Ultrafast cut incident investigation from hours to minutes at OpenAI","summary":"OpenAI engineers use Codex Ultrafast to cut incident investigation from hours to minutes, per OpenAI.","source":"openai_blog","url":"https://openai.com/index/openai-cut-incident-investigation","published":"Mon, 26 Oct 2026 07:00:00 GMT"}]}],"video":{"youtube_id":"eMTV9331zL4","title":"LLM Digest Daily Explainer: Oct 26, 2026","duration_s":115.7}}