https://codecartographer.dev/blog/ Short, cited articles from Team Cartographer on working with coding agents, with sources. en-us Sun, 01 Nov 2026 14:02:44 GMT https://codecartographer.dev/blog/run-a-retrospective-on-your-agent-session https://codecartographer.dev/blog/run-a-retrospective-on-your-agent-session Sun, 01 Nov 2026 12:00:00 GMT Teams have long paused to ask how they worked, not just what they shipped. Here's what Scrum and the Agile principles say, and a small version for one. https://codecartographer.dev/blog/is-a-coding-agent-a-pair-programming-partner https://codecartographer.dev/blog/is-a-coding-agent-a-pair-programming-partner Sat, 31 Oct 2026 12:00:00 GMT Pair programming research found it isn't uniformly beneficial. What that meta-analysis and Anthropic's docs suggest about working with an agent. https://codecartographer.dev/blog/your-agent-session-history-expires https://codecartographer.dev/blog/your-agent-session-history-expires Thu, 29 Oct 2026 12:00:00 GMT Claude Code deletes old session transcripts after 30 days by default. Here's what that means for the record of your build, and a simple way to keep it. https://codecartographer.dev/blog/ironies-of-automation-and-agent-built-code https://codecartographer.dev/blog/ironies-of-automation-and-agent-built-code Tue, 27 Oct 2026 12:00:00 GMT Bainbridge warned in 1983 that automation can erode the skills needed to supervise it. A 2026 trial of AI-assisted coding suggests a similar worry. https://codecartographer.dev/blog/feeling-fast-versus-being-fast https://codecartographer.dev/blog/feeling-fast-versus-being-fast Sun, 25 Oct 2026 12:00:00 GMT Developers in one trial felt faster with AI and measured slower. What that and the planning fallacy suggest about timing your own agent sessions. https://codecartographer.dev/blog/write-the-spec-before-the-prompt https://codecartographer.dev/blog/write-the-spec-before-the-prompt Fri, 23 Oct 2026 12:00:00 GMT Vague requirements are an old problem, and agents don't remove it. What a requirements survey and Anthropic's docs say about writing the spec first. https://codecartographer.dev/blog/when-to-start-a-fresh-session https://codecartographer.dev/blog/when-to-start-a-fresh-session Wed, 21 Oct 2026 12:00:00 GMT Long agent sessions get heavy. What Anthropic's own docs say about context, /clear, /compact and rewind, and a plain rule for when to start over. https://codecartographer.dev/blog/write-down-the-why https://codecartographer.dev/blog/write-down-the-why Mon, 19 Oct 2026 12:00:00 GMT Code shows what you chose, not why. What Nygard and Microsoft's guidance say about decision records, and a ten second habit for agent sessions. https://codecartographer.dev/blog/praise-is-not-evidence https://codecartographer.dev/blog/praise-is-not-evidence Sat, 17 Oct 2026 12:00:00 GMT A confident all-done feels like proof, but it isn't. What research and vendor docs say about checking an agent's work against evidence, not tone. https://codecartographer.dev/blog/reading-your-own-dead-ends https://codecartographer.dev/blog/reading-your-own-dead-ends Thu, 15 Oct 2026 12:00:00 GMT Abandoned branches and rewound sessions are the part of a build nobody keeps. Here's why they're worth reading, and what the docs say about keeping them. https://codecartographer.dev/blog/declaring-done-before-you-start https://codecartographer.dev/blog/declaring-done-before-you-start Tue, 13 Oct 2026 12:00:00 GMT Decide what finished looks like before your first prompt. What Scrum and vendor docs say about done, and how to write a yardstick an agent can check. https://codecartographer.dev/blog/why-a-correction-is-the-most-informative-prompt https://codecartographer.dev/blog/why-a-correction-is-the-most-informative-prompt Sun, 11 Oct 2026 12:00:00 GMT A correction tells you what you meant, what the agent assumed, and where they split. Here's what research and vendor docs say, and how to read yours. https://codecartographer.dev/blog/reviewing-code-you-did-not-write https://codecartographer.dev/blog/reviewing-code-you-did-not-write Fri, 09 Oct 2026 12:00:00 GMT Two studies of code review put understanding the change first. Here's what they found, and what it means when you didn't write the code.