All Articles
What I Stop Doing Before I Hit Claude's Limit — Rebuilding a Day Around the Five-Hour Window
Claude's usage recovers on a five-hour window, but that window doesn't start on the clock — it starts with the first message you send. Here's how I reordered my working day so I stop before the limit stops me.
I Send Images Twenty at a Time — The Day the 21st Image Changed the Rules for the Other 61
Sending 62 images in one request failed with invalid_request_error. The cause was the image count, not the payload size. Here is how I recounted visual tokens as 28px patches and built batches that respect count, dimensions, and payload at once.
The scheduled task that only ran on days my desk was awake
Cowork scheduled tasks run remotely by default, but the moment one needs a local file or app, it runs only on your machine. Here is how the last field in the setup dialog decides that, and the three questions I now answer before creating a task.
Where to Put an Instruction That Keeps Coming Back: Chat, Style, Settings File, or Hook
No comments, please. Do not translate the product name. When an instruction drifts back a few turns later, the fix is usually the location, not the wording. I compared the same sentence across four layers and settled on a rule for which instruction belongs where.
What you keep the agent from walking matters more than what you hand it
Searching one connected Cowork folder for the same word gave three different answers. Here is what the default search silently left out, and how to measure a folder's scan scope once before you hand it over.
Picking Up Where You Left Off: When to Lean on Claude's Memory and When to Ask It to Search
Claude's memory and past-chat search are two different mechanisms. How to tell which one answered you, why project boundaries block searches, why deleting a chat leaves its memory behind, and which plans get what.
When Adding Files Makes Your Claude Project Answer Worse
You upload more material to a Claude Project and the answers get vaguer, not sharper. The cause is overlap, not volume. Here is a small script that measures how much your knowledge files repeat each other, and the three questions I use to decide what stays.
My overnight crawl never came back, and the reason was that WebFetch had no ceiling on how long it would wait
Claude Code v2.1.268 gives WebFetch a 300-second deadline. Here is how an unattended run stalls on a server that never closes its response, why CLAUDE_CODE_WEBFETCH_DEADLINE_MS=0 does not mean 'do not wait', and how I now place two separate ceilings.
inferenceGatewayHeaders still works today. At noon on October 7 it falls back to a default
I read the deprecation notice that started appearing on September 10 as a spelling cleanup. After the cut-off, the old names are not ignored — they fall back to a fail-closed value or a default. Here is the old-to-new map and a script that audits your config in one pass.
When Every Request Fails With Not signed in to the Cloud gateway, Check the Version Before the Config
A Claude Code 2.1.265 regression broke every request for gateway and proxy setups. I use it as an excuse to build a small snapshot of the auth input surface, so the next time something breaks you can tell config from version in seconds.
One plugin.json turns a shelf of skills into something you can switch off as a group
When I sorted my skills by what work they belong to, the set I open almost daily turned out to be 5 percent of the total. Here is how I dropped a .claude-plugin/plugin.json into each group so I could switch a whole shelf off, plus the two loading rules I tripped over on the way.
Using Aqua Voice for Cowork and Everyday Messages — What Changed When I Started Speaking Instead of Typing
I use the dictation app Aqua Voice for briefing Cowork and for messages and email on my iPhone. Notes on how the texture of my writing changed before its speed did, how Mac and iPhone use differs, and what still gets in the way.