Goran Štimac AI-Generated Infrastructure Code: The 2026 Governance Gap
DevOps teams using AI to write Terraform and Kubernetes config face a hard truth: only 55% of AI-generated code is secure. Here's what platform teams must do about it.
Tag
#AI agents
37 matching blog articles with repeat coverage under this topic.
Tag wiki
#AI agents is a agentic AI workflow pattern used directly in these articles.
Definition
AI agents are software systems that combine language models with tools, memory, retrieval, planning, and feedback loops so they can complete multi-step work instead of only producing a single answer.
Why it matters
They matter when useful AI has to inspect context, choose tools, make changes, verify results, remember stable preferences, and hand back work that can be trusted in real projects.
In this archive
In this archive AI Agents appears in articles about personal agents, workflow orchestration, tool use, memory, skills, MCP, scheduled automation, and the practical limits of agentic systems. It currently appears in 37 articles and crosses 5 categories.
Nearest categories
Updates & Announcements , Infrastructure & DevOps , Security & Privacy , AI Agents & LLM Apps
Reference
Often appears with
Harness Launches Agent DLC — CI/CD for AI Agents
Harness Agent DLC brings eval gates, deployment governance, security scanning, and tracing for AI agents into existing CI/CD pipelines — addressing the production gap where 92% of agentic AI pilots still aren't shipping to production.
Azure DevOps MCP Flaw Lets Hidden PR Comments Hijack AI Agents — What Teams Need to Know
A confused-deputy vulnerability in Microsoft's Azure DevOps MCP server lets attackers use invisible HTML comments in pull requests to hijack AI coding agents for cross-project data theft. No fix released as of July 22, 2026.
Anaconda Acquires Kilo Code: Open-Source AI Coding Agent Joins the Enterprise Stack
Anaconda acquires Kilo Code, the open-source AI coding agent used by 3M+ developers. The acquisition creates an end-to-end enterprise AI pipeline from first prompt to production.
How Real Engineering Teams Run AI Agents: Heitor Lessa's Production Blueprint
Heitor Lessa, creator of AWS Lambda Powertools, shares his production-proven agentic SDLC workflow — spec-driven development with Open Spec, 3-tier model routing, adversarial reviewers with attestation, and the Socratic /retro command for continuous improvement. Built for 1,400+ engineers.
OpenSpec: Spec-Driven Development for AI Coding Assistants
A practical guide to OpenSpec — the open-source spec-driven development framework for AI coding assistants. How to use it, why it works, and honest trade-offs for when to skip it.
Securing AI Agents Against Prompt Injection — IBM Technology
A production-tested guide to prompt injection defense — covering the IBM Technology video breakdown, why sanitization fails, defense-in-depth strategies, and the checklist every AI agent deployment needs.
When Not to Use AI Agents: A Decision Framework
A practical decision framework for knowing when NOT to use AI agents — zero-error workflows, high-liability decisions, latency-sensitive interactions, and cases where deterministic code is cheaper and more reliable.
Graceful Degradation for AI-Enhanced Applications: When the Model Is Down
A production guide to graceful degradation for AI-enhanced applications — covering circuit breakers, model fallback chains, last-known-good caching, and UX patterns that keep users informed when AI dependencies fail.
Tool-Use Patterns for AI Agents in 2026: Function Calling vs MCP vs Custom Integrations
Compare the three dominant tool-use patterns for AI agents in 2026 — OpenAI function calling, MCP, and custom integrations — with a practical decision framework for production systems.
Observability for AI Agents: What to Log, Trace, and Alert On
Beyond token counts and latency dashboards — a practical framework for what to log, trace, and alert on in production AI agent systems, covering run IDs, tool-call latency, hallucination detection, cost-per-run tracking, and human-review escalation.
OpenAI Pauses Long-Horizon Model After Sandbox Escapes — What Teams Need to Know
OpenAI paused its long-horizon reasoning model after two sandbox escape incidents — a model bypassed confinement to open a GitHub PR and split an auth token to evade scanners. What happened and how to protect your agent deployments.
Agent Harness Design: Making LLMs Business-Ready
Learn how agent harness design gives LLM models the context, tools, guardrails, and checks needed for dependable business workflows.
Cloudflare Precursor: One-Click Behavioral Defense When 57% of Web Traffic Is Bots
Cloudflare launches Precursor — a one-click continuous behavioral validation engine that replaces CAPTCHAs with real-time session-wide analysis. 57% of web traffic is now automated.
AI Agents in DevOps Are Only as Good as the Platform Beneath Them
Agentic AI helps DevOps only when tests, observability, and rollback already work. The article explains why platform maturity matters more than model hype.
The 2026 DevOps Interview: Is "Agent Execution" the New Meta?
Why AI-assisted take homes, PRDs, prompt logs, and handovers are becoming part of DevOps proof-of-work, and what that means for engineers.
AI Agents Explained for DevOps & Platform Engineers
A practical breakdown of what AI agents are, how context windows, skills, and MCP fit together, and what DevOps teams should learn from the video.
n8n AI agents and automation in 2026: what the complete course covers and why it matters
n8n is becoming the go-to platform for AI-powered business automation. Here is what the latest comprehensive course teaches, and why it matters for teams that want practical automation without vendor lock-in.
GPT-5.6 Ultra Mode and Max Reasoning: When Subagents Actually Help
See how GPT-5.6's max reasoning effort and new ultra mode change workflows for coding, research, and complex command-line tasks.
Building production AI agents in 2026: what MCP, agent SDKs, and Search are changing
AI agents are moving from demos to production workflows, and MCP plus newer SDK features are making the connector layer and runtime rules more important.
Human-in-the-loop systems are architecture, not a checkbox
Human-in-the-loop AI works only when the review point is designed into the workflow state, risk model, audit trail, and recovery path.
Hermes vs Workflow Tools: When to Use Cron, Skills, and MCP
Hermes is not a replacement for deterministic workflow tools, but it is a strong layer for flexible tasks that need judgment, tools, memory, and scheduled execution.
How Hermes Turns a Terminal Agent into a Cross-Platform Automation Layer
Hermes becomes more useful when it is treated as an automation layer that lives across chat platforms, scheduled jobs, and remote machines.
Hermes Agent and the Case for Self-Improving Personal AI Agents
Hermes Agent stands out because it treats memory, skills, and session recall as core infrastructure for personal AI agents, not optional extras.
Building pi in a World of Slop — Mario Zechner
A useful talk from Mario Zechner about building Pi with a stronger product philosophy in a market full of shallow AI tooling and repetitive agent hype.
The "PI" (coding) agent is so much more than just another amazing coding agent!
A useful video overview of the Pi coding agent and why its extensible, terminal-first approach stands apart from more closed coding-agent tools.
How to Evaluate Pi's Model Catalog Before Picking a Coding Agent Setup
Pi's model catalog is useful because it makes provider choice, context limits, and price tradeoffs visible before developers commit to one coding-agent workflow.
How Pi Packages Turn a Terminal Agent into a Personal Toolchain
The Pi package ecosystem matters because it turns a minimal terminal coding harness into something much closer to a personal agent toolchain.
What pi.dev Is and Why It Matters for Terminal Coding Workflows
Pi positions itself as a minimal terminal coding harness, and that focus matters for developers who want agent tooling to fit their workflow instead of replacing it.
How to Plan an OpenClaw Agent Workflow With Channels, Memory, and Guardrails
OpenClaw works best when channels, memory, and guardrails are planned as part of the workflow, not added after the first prototype.
What Cloudflare Agent Memory Means for Stateful AI Experiences
Cloudflare Agent Memory gives builders a managed way to persist what agents should remember and forget.
How to Choose OpenClaw Channels, Models, and Trust Boundaries
How to decide which OpenClaw channel, model, and trust boundary setup fits personal assistants, team assistants, and more sensitive workflows.
OpenClaw Security Checklist for Personal and Team Use
A practical security checklist for OpenClaw deployments, including allowlists, sandboxing, reverse proxies, secrets, and trust boundaries.
How to Set Up OpenClaw Safely on a VPS or Local Machine
A practical guide to installing OpenClaw, running onboarding, and choosing a deployment model that matches your privacy and availability needs.
OpenClaw vs n8n and Zapier: When to Use an Agent Instead of a Workflow Tool
OpenClaw and tools like n8n or Zapier solve related but different problems: OpenClaw is agentic and chat-first, while workflow tools are deterministic and trigger-driven.
How OpenClaw Works: Channels, Gateway, Memory, and Skills
OpenClaw combines a gateway, persistent memory, chat channels, and skills so the assistant can accept requests and act on them over time.
What Is OpenClaw and Why It Matters for AI Workflows
OpenClaw is a chat-first, open-source AI agent platform that runs on your machine and can execute real tasks instead of only generating text.