https://usewalkie.com/blog/ Measured numbers on tokens, screenshots, and what your coding agent can actually read. en-us https://usewalkie.com/blog/why-your-agent-cant-see-what-it-built/ https://usewalkie.com/blog/why-your-agent-cant-see-what-it-built/ An AI agent's world is text — it can't see what its code renders unless a tool hands it an image, and even then it needs to know what to look for. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/why-claude-max-runs-out/ https://usewalkie.com/blog/why-claude-max-runs-out/ Claude usage limits are token-based, not message-based — long visual sessions compound fast, and the fixes cost nothing. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/what-is-mcp/ https://usewalkie.com/blog/what-is-mcp/ MCP is Anthropic's open standard letting an AI agent call real tools and data instead of only generating text, one server per connected system. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/what-is-a-context-window/ https://usewalkie.com/blog/what-is-a-context-window/ A context window is the total text a model holds per turn; filling a 200,000-token window can cost 40% of it before a fix even starts. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/what-a-claude-code-session-costs/ https://usewalkie.com/blog/what-a-claude-code-session-costs/ A line-item breakdown of a two-hour Claude Code session shows screenshots and recordings, not code, driving the biggest single cost. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/video-screenshots-or-words/ https://usewalkie.com/blog/video-screenshots-or-words/ A worked comparison of what a frame-dumped video, screenshots, and spoken words each cost an AI coding agent to read, using Anthropic's formula. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/vibe-coding-workflow/ https://usewalkie.com/blog/vibe-coding-workflow/ The vibe coding workflow is describe, generate, look, react — the look-and-react half burns the most time and gets almost no dedicated tooling. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/vibe-coding-tools-2026/ https://usewalkie.com/blog/vibe-coding-tools-2026/ The 2026 vibe-coding stack has four jobs, agents, voice input, review, and testing, with prices given only where they're actually verified. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/verifying-ai-built-ui-checklist/ https://usewalkie.com/blog/verifying-ai-built-ui-checklist/ Twelve checks for what AI coding agents most reliably miss, why they miss each one, and the fastest way to trigger it to verify the fix. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/screen-recording-for-claude-code/ https://usewalkie.com/blog/screen-recording-for-claude-code/ Four ways to show Claude Code your screen, compared on token cost and what each one loses on the way to the agent. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/reviewing-code-you-didnt-write/ https://usewalkie.com/blog/reviewing-code-you-didnt-write/ The order that works: data shape, then boundaries, then logic. Top-to-bottom reading misses what AI-generated code gets wrong — and what to grep for. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/review-ai-generated-code/ https://usewalkie.com/blog/review-ai-generated-code/ How to review AI-generated code: what the agent already checked, what only you can catch, and a loop that closes the gap. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/plan-execute-verify-loop/ https://usewalkie.com/blog/plan-execute-verify-loop/ Plan and execute are easy to automate. Verification is where the quality is — and it's the one step an AI agent can't fully do by itself. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/loom-video-to-ai-coding-agent/ https://usewalkie.com/blog/loom-video-to-ai-coding-agent/ Loom's AI reads the transcript, not the pixels — so a Loom link never gets your coding agent the screen you pointed at. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/give-feedback-to-claude-code/ https://usewalkie.com/blog/give-feedback-to-claude-code/ Good feedback to Claude Code names the location, the expectation, and the gap between them — most reports name only the gap, and the fix misses. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/describe-a-visual-bug-to-ai/ https://usewalkie.com/blog/describe-a-visual-bug-to-ai/ 'The spacing looks off' gives an agent nothing to fix. Here's what to say instead, with a translation table of real vague-to-specific examples. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/cut-claude-code-token-costs/ https://usewalkie.com/blog/cut-claude-code-token-costs/ Eleven ways to cut Claude Code token costs, ranked by tokens saved per minute of effort. Ten are free; one costs money. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/claude-code-token-costs/ https://usewalkie.com/blog/claude-code-token-costs/ A 1080p screenshot costs 2,691 tokens by Anthropic's own formula. Here's where the rest of a Claude Code session's tokens actually go. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/agent-feedback-tools-compared/ https://usewalkie.com/blog/agent-feedback-tools-compared/ A fair, priced comparison of Walkie, Clipy, Cobalt Capture, Vibeshots, Vibe Annotations, talkthrough-mcp, Jam.dev and plain screenshots. Roberto Ercole Wed, 05 Aug 2026 00:00:00 GMT https://usewalkie.com/blog/why-ai-cant-watch-screen-recordings/ https://usewalkie.com/blog/why-ai-cant-watch-screen-recordings/ Claude's API accepts images, not video. What actually happens when you frame-dump a screen recording for your agent — and the token cost. Roberto Ercole Wed, 29 Jul 2026 00:00:00 GMT https://usewalkie.com/blog/what-is-review-md-bundle/ https://usewalkie.com/blog/what-is-review-md-bundle/ A plain look at Walkie's REVIEW.md bundle format — what's inside it, how big it is, and why your coding agent can read it but not your screen recording. Roberto Ercole Wed, 29 Jul 2026 00:00:00 GMT https://usewalkie.com/blog/walkie-vs-clipy-vs-loom-ai-screen-review-tools/ https://usewalkie.com/blog/walkie-vs-clipy-vs-loom-ai-screen-review-tools/ A fact-checked comparison of Walkie, Clipy, Loom, and other screen-to-AI review tools on price, tokens, and features. Roberto Ercole Wed, 29 Jul 2026 00:00:00 GMT https://usewalkie.com/blog/screenshot-token-cost-ai-coding-agent/ https://usewalkie.com/blog/screenshot-token-cost-ai-coding-agent/ A measured, sourced breakdown of screenshot token cost for Claude and other coding agents, plus what to send instead. Roberto Ercole Wed, 29 Jul 2026 00:00:00 GMT https://usewalkie.com/blog/give-ai-coding-agent-visual-feedback-without-burning-tokens/ https://usewalkie.com/blog/give-ai-coding-agent-visual-feedback-without-burning-tokens/ The cheap way to show Claude Code or Cursor a UI bug: narrate and point instead of pasting screenshots. Here's the token math and the workflow. Roberto Ercole Wed, 29 Jul 2026 00:00:00 GMT https://usewalkie.com/blog/claude-image-token-calculation-explained/ https://usewalkie.com/blog/claude-image-token-calculation-explained/ The exact formula Claude uses to price an image in tokens, worked through on a real 1920x1080 screen and a measured test screenshot. Roberto Ercole Wed, 29 Jul 2026 00:00:00 GMT