Chisato · · 4 min read What Is Context Rot in LLMs?
Context rot is the drop in an LLM's accuracy and reliability as the amount of text in its context window grows, even when the window technically fits it.
#AI
#LLMs
#Agents
Topic
#Agents
27 posts tagged “Agents”.
Chisato · · 4 min read 
Chisato · · 5 min read Atria Dawn Preview: Shanghai AI Lab's 744B Open Agent
Shanghai AI Lab quietly released Atria Dawn Preview, a 744B MoE agentic model under MIT license built on GLM-5.2. Specs, benchmarks and the caveats.
#AI
#Open Source
#Agents
Chisato · · 6 min read OpenAI Agents API: Codex Harness Now in Public Beta
OpenAI opened its Agents API in public beta on Sept 10, putting the managed Codex harness behind one API call. What it does, how sandboxes work, and pricing.
#AI
#OpenAI
#Agents
Chisato · · 4 min read McKinsey State of AI 2026: Agents Scale, Trust Lags
McKinsey's 2026 survey finds enterprises scaling AI agents from 27% to 40% of firms, with a third skipping software purchases to build in-house — but governance trails.
#AI
#Agents
#Enterprise
Chisato · · 5 min read LLM Observability: Tracing AI Agents in Production
LLM observability traces every prompt, tool call, and token spent across an agent's run, turning an opaque chain of model calls into something debuggable.
#AI
#LLMs
#Agents
Chisato · · 5 min read AI Agent Sandboxing: How Agents Run Code Safely
Agent sandboxing isolates the code an AI agent executes from the host system, limiting what a compromised or misbehaving agent can actually reach.
#AI
#Security
#Agents
Chisato · · 7 min read AISI Report: AI Agents Took Rogue Actions in Cyber Tests
The UK's AI Security Institute found agents built on Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol took 19 unsanctioned actions against real targets.
#AI
#Security
#Agents
Chisato · · 6 min read Grok Voice Think Fast 2.0: Pricing, Specs, Default Date
xAI's Grok Voice Think Fast 2.0 becomes the default grok-voice-latest on Aug 5, with an 82.9% speech-quality score and $0.08/min pricing. What changed.
#AI
#LLMs
#Agents
Chisato · · 5 min read AI Agent Memory: Short-Term vs Long-Term Context
How AI agents remember: short-term memory bound by the context window versus long-term memory persisted in external storage like a vector database.
#AI
#LLMs
#Agents
Chisato · · 4 min read The ReAct Pattern: How AI Agents Reason and Act
ReAct interleaves an LLM's reasoning with tool calls and their results, letting an agent adjust its plan after each observation instead of reasoning blind.
#AI
#Agents
#LLMs
Chisato · · 6 min read MCP 2026-07-28 Update: Stateless Protocol, New Auth
The Model Context Protocol dropped sessions, killed the init handshake, and rewrote authorization in its biggest spec change yet. What changes for AI agents.
#AI
#MCP
#Agents
Chisato · · 4 min read What Is an LLM Router?
An LLM router sends each request to the cheapest or fastest model that can handle it, instead of routing every call to one model regardless of difficulty.
#AI
#LLMs
#Agents
Chisato · · 4 min read What Is a Multi-Agent System? AI Agents Working Together
A multi-agent system splits a task across several specialized AI agents that coordinate instead of one agent doing everything. How they're structured.
#AI
#LLMs
#Agents
Chisato · · 6 min read ARD: Big Tech's Agent Standard vs Anthropic's MCP
ARD vs MCP: Big Tech's new agent-discovery standard takes aim at Anthropic's protocol. What ARD does, who backs it, and how the two actually differ.
#AI
#Agents
#Enterprise
Chisato · · 6 min read Grok 4.5 Launch: Pricing, Benchmarks, Opus Claim
SpaceXAI's Grok 4.5 ships as an 'Opus-class' coding model at $2/$6 per million tokens. Benchmarks vs Opus 4.8, token efficiency, and where it fits.
#AI
#LLMs
#Agents
Chisato · · 6 min read Meta AI Reset: Zuckerberg Admits Progress Stalled
At a July 2 town hall, Mark Zuckerberg told staff Meta's AI agent work 'hasn't really accelerated' — months after 8,000 layoffs and a costly reorg. What it signals.
#AI
#Meta
#Agents
Chisato · · 6 min read JADEPUFFER: First Agentic Ransomware Attack Explained
Sysdig documented JADEPUFFER, the first ransomware run end-to-end by an AI agent — how it exploited Langflow, encrypted a database, and why it matters.
#Security
#AI
#Ransomware
Chisato · · 5 min read Claude Science: Anthropic's AI Workbench for Labs
Anthropic launched Claude Science, an agentic research workbench with 60+ skills for genomics, chemistry, and more. What it does and who it's for.
#AI
#Anthropic
#Agents
Chisato · · 4 min read Build Your Own AI Agent in 100 Lines of Python
Build a real AI agent from scratch — no framework. Just the Anthropic API, a tool-use loop, and two tools the model can call to explore your files.
#AI
#Agents
#LLMs
Takina · · 2 min read Cloudflare Builds Out Its Platform for AI Agents
Cloudflare's June updates target AI agents: longer-lived Durable Objects, tokenless temporary deploys, PlanetScale databases, and a private bot protocol.
#Cloud
#Developer Tools
#Agents
Chisato · · 4 min read Claude Fable 5: Anthropic's Most Capable Model Yet
Anthropic's Claude Fable 5 is its most capable model yet, built for long-horizon, autonomous agent work. Here's what's new, what it costs, and when to use it.
#AI
#Claude
#Anthropic
Chisato · · 7 min read Omnigent Quickstart: Orchestrate Claude Code, Codex & Pi
Hands-on with Omnigent, Databricks' open-source meta-harness: install it, run your first agent, swap harnesses, and add cost and approval policies.
#AI
#Agents
#Databricks
Chisato · · 6 min read Databricks Omnigent: Open-Source Meta-Harness for AI Agents
Databricks open-sourced Omnigent, a meta-harness that unifies Claude Code, Codex, Cursor, and Pi in one layer for composition and control.
#AI
#Agents
#Databricks
Chisato · · 2 min read xAI's Grok 4.3 Arrives as a Budget Frontier Model
xAI's Grok 4.3 hit Amazon Bedrock as the cheapest US frontier reasoning model, while the 6-trillion-parameter Grok 5 slips. Here's where xAI stands in 2026.
#AI
#LLMs
#Agents
Chisato · · 5 min read Model Context Protocol (MCP), Explained
The Model Context Protocol (MCP) is the USB-C of AI — one open standard that lets any model plug into your tools and data. How it works and why it won.
#AI
#Agents
#Developer Tools
Chisato · · 4 min read What Is an AI Agent? Goals, Tools, and the Loop
An AI agent is an LLM-powered system that pursues a goal across steps — planning, calling tools, observing results, and repeating until the job is done.
#AI
#Agents
#LLMs
Chisato · · 3 min read What Is Letta AI? Stateful Agents with Real Memory
Letta (formerly MemGPT) builds stateful AI agents with long-term memory that persists across sessions. Here's what Letta is and how its memory model works.
#AI
#Agents
#Open Source