The post Docker is weird explains containers in under 500 words, with headings that make claims.
Badges
Badges come from GitHub once the author claims this feed.
Recent posts
every link opens the author’s post-
Why 'Docker is weird' works as an explainer
-
What I'd copy from Cloudflare's KV cache incident report
Cloudflare's report on a KV cache leak in Workers AI puts the count of affected responses first and admits what the team knew before launch.
-
Agent docs don't agree on what a harness is
An essay on vocabulary. Runta's benchmark calls Codex CLI a harness, Anthropic uses the word for its hosted Managed Agents and LangChain for its Deep Agents package.
-
Notes on Chrome's Modern Web Guidance for coding agents
Chrome shipped more than 200 guides written for coding agents.
-
I went looking for the citation field in a proposed agent memory format
A proposal for a shared agent memory format lists five questions a record must answer.
-
Applying Google's pager rule to a build that is always red
A team that let agents write every line found its build frequently red.
-
I read SigNoz's post on AI deskilling SREs and thought about runbooks
SigNoz argues that AI is deskilling SREs.
-
Thoughts on Ghostty's hand-crafted release notes
Ghostty's release notes say people wrote them.
-
What I'd still write by hand in a codebase agents can explain
A Reddit post said nobody understands the code after a year of agents.
-
Why I rebuilt my best-LLMs-for-coding page
My March ranking of coding models rested on one benchmark score.
-
What Fern's agent traffic report made me change in my llms.txt article
Fern says agents are now over half of the requests to the docs it hosts.
-
Runbook Template and Example: Steps an On-Call Agent Can Follow
A Markdown runbook template where each step has a command, its expected output and a branch, with a tested Postgres example and a checker to run first.
-
AEO vs GEO vs SEO: What Changes for Developer Docs
What AEO, GEO and SEO each mean, what Google and OpenAI document about AI search, and six checks that decide whether developer docs can be found and quoted.
-
GPTBot, ClaudeBot and OAI-SearchBot: AI Crawlers in robots.txt
Which robots.txt rules block AI training crawlers like GPTBot and which block AI search, plus what 12 developer documentation sites allow.
-
API Documentation: What to Include, With a Tested Reference Example
What API documentation includes, page by page, and a complete reference entry for one endpoint that a script checks against the running service.
-
llms.txt: What It Is, What It Does, and Four Real Files Checked
What an llms.txt file is, what the proposal requires, what Google, Chrome and server logs say about its use, and four real files checked against the proposal.
-
Claude Code Subagents: What Each One Sees, With a Tested Example
What a Claude Code subagent starts with and when to pick one over a skill or hook, with a link-checker subagent tested in Claude Code 2.1.308.
-
CLAUDE.md and AGENTS.md: Which File Loads, and a Template
Which coding agent reads CLAUDE.md and which reads AGENTS.md with default settings, tested in four repositories, plus a copyable AGENTS.md template.
-
MCP Server Tutorial: Build a Docs Search Server in Python
Build an MCP server in Python, test it without a model, and see how numbering a tool's output changed whether Claude cited the right line in a test run.
-
Design Doc Template: Sections, a Filled Example and a Review Check
A design doc template in Markdown with goals, non-goals and a required "do nothing" alternative, a filled example, and a checker to run before review.
-
Quick Start Guide Template: Write a Quickstart Developers Can Run
A quick start guide template for developer products, with a filled example against the GitHub REST API and a checklist to run before you publish.
-
Release Notes Template: Copyable Markdown With a Pre-Publish Check
A release notes template in Markdown that puts the upgrade decision first, with a filled SDK example and a checker that fails a note missing its rollback line.