Tag: agents

All the articles with the tag "agents".

• Agent Sprawl Is a Skills Problem

  6 Jul, 2026

  I had hundreds of skills scattered across agent harnesses, repos, and drafts folders. No single view of what existed, what was current, or what was actually wired in. The market is hitting the same wall.

• The Cross-App Workflow Cost $5.44. Here's How We Got It to $0.17.

  1 Jun, 2026

  A hybrid agent running browser research + desktop spreadsheet creation cost $5.44 baseline. Optimizing each leg independently — Playwright for the browser, CLI-first prompt for the computer use desktop leg — brought it to $0.17. The browser leg did most of the damage.

• The Test Isn't Whether the Agent Picks the Right Tool. It's What Happens When the Tool Fails.

  31 May, 2026

  I built a hybrid agent expecting the routing to be the demo. The actual demo turned out to be what happened when a tool call failed and the outer model retried with a different approach — twice in two runs.

• I Had a Working Agent at $3.41 a Run. Here's Where the Money Was Actually Going.

  30 May, 2026

  Same task, different harness configurations. Caching the inner loop saved 29%. Convergence rules saved 54%. Eliminating the LLM from the browser loop entirely dropped cost from $3.41 to $0.03. The harness is where the margin lives — and the harness goes all the way down.

• API → Connector → Browser → Computer Use → Human: A Cost-Justified Hierarchy for Agent Tooling

  29 May, 2026

  Computer use is 45× more expensive than a structured API for the same task. Eight harness configurations across browser, desktop, and hybrid showed costs from $0.03 to $5.44 on the same task — the hierarchy, the numbers, and why implementation beats model choice.

• Composer 2.5 Earned a Daily-Driver Slot — After Two Days of Real Wiring Work

  21 May, 2026

  First impressions of Cursor Composer 2.5 in the agents UI: integration work across repos, 328 requests and ~202M tokens on the included pool, 95%+ cache hits — not a review, a usage report.

• The AI Chief of Staff Is a Fleet, Not an Agent

  15 May, 2026

  The AI Chief of Staff pattern is everywhere in May 2026 — and every implementation makes the same mistake. They build one agent that generates a morning brief. A real chief of staff is a fleet.

• The Enterprise MCP Pattern: Proxy, Aggregate, Host

  13 May, 2026

  There are 58 public MCP servers in the awslabs catalog covering cloud infrastructure, databases, AI/ML, and ops tooling. Here's the enterprise pattern: proxy them locally, then host the proxy behind a gateway so every developer in your org gets them without setup.