
The Multi-Feature Request: Batching Work for AI Efficiency
Four features, one prompt, zero clarifications. How to structure multi-feature requests that don’t need hand-holding.

Four features, one prompt, zero clarifications. How to structure multi-feature requests that don’t need hand-holding.

No Photoshop. No local image editing. Just CDN transforms that turn stock photos into moody Hong Kong cinema.

One sentence launched an entire content capture system. Breaking down what made a vague-sounding prompt surprisingly effective.

The button was visible. The selector found it. The click did nothing. An invisible modal overlay was stealing every interaction.

OpenCode failed with ‘ProviderModelNotFoundError.’ The model ID existed—just not spelled the way my config spelled it.

My MCP server config worked in Claude Desktop. It failed validation in OpenCode. Same protocol, different schemas.

I completed the browser auth flow. The script still said ‘Auth required.’ The shell wrapper and TypeScript implementation disagreed about what ‘authenticated’ meant.

The gateway failed with ‘missing env var.’ I had the token—just under the old project name.

My test script died silently after the first passing test. The culprit: ((var++)) combined with set -e.

My Hugo blog showed zero posts. The content was there—just pointed at a user who doesn’t exist on this machine.