Topic
How teams write, fund and maintain documentation, and the contrast between human-facing docs and agent-facing instructions.
Claude Code's configuration-debugging page says nested memory files load on demand through the Read tool, so a report that the agent ignored the rules starts with /context and the question of whether the file was in the window at all.
Reality
On @forge/cli 13.4.0 the linter runs inside forge deploy and blocks it, so a clean run looks like a passed check. One request helper is enough to leave a missing scope to surface as a 401 in production.
Reality
One developer-docs maintainer routes his assistant's failed answers into a job that drafts a change and files a pull request. The trigger reads the assistant's own reply, so a confident wrong answer still needs a human to flag it.
Reality
Vendor pages for Claude Code, Cursor, Codex, VS Code and Windsurf, re-checked on 2026-09-12, show that all five resume a previous session. The complaint about agents forgetting points at the other three questions.
Reality
A 558-file corpus puts gotchas last of nine categories at 13.6%, and the same study reports 32 to 38% recall on that slot, so the sturdier evidence for scarcity is a hand read of 120 pitfall entries.
Reality
A four-layer model of how coding agents choose an SDK pushes the work past documentation into llms.txt and MCP servers. The search layer comes with a measured trigger rate from Vercel; the layers below it come with none.
Reality
Teams now maintain agent context files line by line because tokens are metered. The same teams left the human wiki to rot for a decade.
Reality