Anatoliy Kolodkin - The LGTM (Page 22)

The LGTM

Home
Agentic Coding
Claude Code
Codex

Sign in Subscribe

Anatoliy Kolodkin

Claude Reviewing Claude Is Structurally Circular — Without Executable Specs, AI Code Review Just Echoes Its Own Mistakes

Claude Reviewing Claude Is Structurally Circular — Without Executable Specs, AI Code Review Just Echoes Its Own Mistakes

The most common response to AI-generated code quality problems is deploying an AI reviewer. A new paper argues that response is structurally circular when you don't have executable specifications in place. The argument is not that AI code review is bad — it's that when the reviewing

30 Mar 2026 1 min read

Manager Agents That Hire and Fire Their Own Workers: The First Framework Built for Continuous (Not One-Shot) Software Development

Manager Agents That Hire and Fire Their Own Workers: The First Framework Built for Continuous (Not One-Shot) Software Development

Nearly every agentic coding framework in production today was built to solve one-shot tasks: hand the agent a well-scoped ticket, receive a patch. That architecture breaks down the moment software development is what it actually is — iterative, multi-day, and evolving. Requirements shift mid-project. New complexity surfaces that wasn't

30 Mar 2026 1 min read

Strong Manager + Cheap Worker = Top-Model Performance at a Fraction of the Cost

Strong Manager + Cheap Worker = Top-Model Performance at a Fraction of the Cost

What happens when you pair an expensive reasoning model with a cheap execution model to solve software engineering tasks? A new paper studying the ManagerWorker architecture provides the clearest empirical answer yet — and the headline finding is immediately actionable. A strong manager directing a weak worker achieves a 62% resolve

30 Mar 2026 1 min read

Your Agent's PRs Get Rejected Because They Ignore Your History: Learning Organic Commits via Repository Memory

Your Agent's PRs Get Rejected Because They Ignore Your History: Learning Organic Commits via Repository Memory

Coding agents have made remarkable strides on benchmark leaderboards, yet teams deploying them in production keep running into the same quiet rejection: the AI-generated pull request technically works but doesn't fit. It duplicates an internal API that a teammate wrote two years ago. It ignores the project'

30 Mar 2026 2 min read

CVE-2026-32974: OpenClaw Feishu Webhook Vulnerability Advisory

CVE-2026-32974: OpenClaw Feishu Webhook Vulnerability Advisory

RedPacket Security has published an advisory for CVE-2026-32974, a high-severity vulnerability in OpenClaw's Feishu webhook endpoint. The flaw allows an attacker to send crafted requests directly to the endpoint with low complexity and no prior privileges required — a concerning combination for any deployment with Feishu integration enabled. A

29 Mar 2026 1 min read

The Rise of the 'Claws': Understanding OpenClaw's Security Risks and the New Agentic Era

The Rise of the 'Claws': Understanding OpenClaw's Security Risks and the New Agentic Era

The Investing in AI newsletter delivers a deep-dive analysis of the broader OpenClaw ecosystem, examining everything from Meta's acquisition of Moltbook to mounting security concerns around agent impersonation. The piece argues that OpenClaw has created a new kind of viral loop — where the spectacle of autonomous agents in

29 Mar 2026 1 min read

OpenClaw Is Taking the Tech World by Storm — Here's What You Need to Know

OpenClaw Is Taking the Tech World by Storm — Here's What You Need to Know

Yahoo Finance chronicles OpenClaw's whirlwind journey from niche developer tool to mainstream AI phenomenon. The profile traces the arc from creator Steinberger's move to OpenAI, through Nvidia featuring him on stage at GTC, to the now-iconic lobster-claw logo becoming actual headgear at tech conferences. It'

29 Mar 2026 1 min read

OpenClaw: China's Latest Tech Obsession Could Be a Game Changer

OpenClaw: China's Latest Tech Obsession Could Be a Game Changer

CNN is calling OpenClaw China's latest tech obsession — and the coverage paints a vivid picture of the open-source agent framework's cultural impact on the other side of the Pacific. From lobster-themed developer meetups to claw-machine-inspired branding popping up at tech conferences, China's developer community

29 Mar 2026 1 min read

The Control Layer: Why Agentic AI Frameworks Are The Next Big Thing

The Control Layer: Why Agentic AI Frameworks Are The Next Big Thing

A market analysis from Inc42 puts into sharp focus a thesis that's been building quietly for the past year: the AI framework layer — orchestration platforms like LangChain, LlamaIndex, LangGraph, AutoGen, Google Vertex AI Agent Builder, and AWS Agents for Bedrock — is becoming the new operating system for enterprise

29 Mar 2026 1 min read

Google ADK 2.0 Alpha 1: Building Deterministic AI Agents with Graph-Based Workflows

Google ADK 2.0 Alpha 1: Building Deterministic AI Agents with Graph-Based Workflows

Google's Agent Development Kit team has released ADK 2.0 Alpha 1, and the centerpiece of the release is a fundamental architectural shift: Deterministic Workflows modeled as Graphs. The core problem the team set out to solve is probabilistic unreliability — the inherent unpredictability of LLMs when deployed in

29 Mar 2026 1 min read

EU AI Act + LangChain: What You Actually Need to Build Before August 2026

EU AI Act + LangChain: What You Actually Need to Build Before August 2026

The EU AI Act's high-risk enforcement deadline is 126 days away — August 2, 2026 — and a new technical breakdown from the DEV Community makes clear that most production LangChain setups are not ready. The gap isn't about intent; it's architectural. What the Act actually

29 Mar 2026 1 min read

The PRP Playbook: How to Engineer Context So Your AI Coding Assistant Never Misses Your Architecture Again

The PRP Playbook: How to Engineer Context So Your AI Coding Assistant Never Misses Your Architecture Again

Every team running AI coding assistants hits the same wall within a few weeks: the AI generates code that misses the architecture, ignores testing standards, and breaks the deployment pipeline. A new breakdown of the rapidly-growing context-engineering-intro repository by coleam00 argues that the fix isn't a better prompt

29 Mar 2026 1 min read

← Newer Posts Page 22 of 46 Older Posts →

The LGTM © 2026

Sign up

Powered by Ghost