OpenAI Decisions API: Uses and Enterprise Adoption Considerations
How OpenAI Decisions API returns probabilities, choices, and scores, with text and image use cases, pricing, error evaluation, data retention, and execution controls.
AI Agents
17 posts with the tag “AI Agents”
AI Capabilities and Agent Coordination: Lessons from the OpenAI and Gemini Intrusions
Compare the OpenAI and Gemini intrusions using public sources. Examine what exploration and coordination achieved, what remains uncertain, and how information sharing, permissions, and stopping affect business use.
WebMCP Explained: How AI Works with a Live Web Page
A beginner-friendly explanation of why WebMCP was proposed, its two ways to describe page features, how it differs from adjacent technologies, and how to test it safely.
What Is AdCP? AI Ad Creation, Approval, Delivery, and Corporate Responsibility
Using the AdCP Creative, Governance, and Trust specifications, this article separates information the protocol can carry from the checks and management responsibilities organizations retain.
Agent Observability Without Centralizing Conversation Content
A design for monitoring long-running agent behavior through minimal safety signals without centralizing conversations and documents.
Verification-First Agents: Validating AI Answers Through Evidence
An introduction to Verification-First Agent design, which links each AI claim to evidence and an independent check before the answer informs a decision or action.
What Is OpenMontage? Managing AI Video Production in the Enterprise
Using OpenMontage as an example, this article explains how to organize AI video production stages, responsibilities, approvals, records, and enterprise adoption checks.
What Is Palantir Ontology? Connecting Business Data, Decisions, and Actions
Understand Palantir Ontology through its distinction from general ontology, Objects, Links, Functions, Actions, permissions, operation records, its role alongside knowledge graphs and RAG, and criteria for scoping a pilot.
What Is Hermes Agent? Differences, Use Cases, and Extensions
A research-based overview of Hermes Agent covering framework differences, use cases, extensions, and considerations for organizational use.
What Is SkillOps? Agent Skill Operations and Maintenance at Google
Using Google's published approach to building, testing, and scaling Agent Skills, this article explains SkillOps maintenance, comparative evaluation, and ownership.
Choosing AI Agent Output: HTML or Markdown
Using the Claude Code team's HTML workflow as a starting point, this article explains when to use Markdown, HTML, XML, and JSON for AI agent outputs.
Self-Healing Loops for Coding Agents: Verification, Stopping, and Human Handoff
A practical guide to coding-agent self-healing loops based on 2026 research and official implementations from OpenAI, Google, and GitHub, covering layered verification, stop conditions, permission boundaries, and human approval.
Claude Dynamic Workflows: Architecture and Parallel Task Design
A plain-language guide to Dynamic Workflows in Claude Code: how multiple AI workers divide a job, which tasks fit, and how to plan outputs, limits, costs, and resumption.
How to Preserve AI Workflow Decisions in lessons.md
How to record an AI workflow problem, cause, resolution, and prevention in lessons.md as a decision history that later work can verify.
Approval Gates for AI Agent Build Commands
A record of how a production build command ran while I was asking AI to make a separate change, and the approval-required rule I put in place afterward.
What Is Harness Engineering? A Three-Layer AI Development System
AI gives different answers to the same question, and it does not remember previous sessions. This article explains how I use the term Harness Engineering for this site's AI operation practices.
What Work to Delegate to AI: Three Decision Criteria
A pre-delegation risk check for AI tasks, using reversibility, scope of impact, ease of verification, and the human decisions that should remain under human control.