
CLAUDE.md and AGENTS.md: What Coding Agents Read
How Claude Code and Codex load project instruction files, what belongs in them, and a worked AGENTS.md for a small Next.js repo.
Start reading 11 min read106 posts, newest first

How Claude Code and Codex load project instruction files, what belongs in them, and a worked AGENTS.md for a small Next.js repo.
Start reading 11 min read
Install Marker, convert a PDF, read the output, then do the same through the Datalab API. Every command comes from the official READMEs and docs.

Which documents a person should check, how many to spot-check, how many reviewer hours you need, and how corrections make the system better.

Per-page prices for Datalab, Extend, Azure, Google and LLM APIs as of September 2026, and a worked example showing why human review is most of the bill.

What changed when document parsing moved from OCR to AI models that read the whole page, with dated examples from Datalab, Extend, Azure and Google.

Thirty document AI terms, from OCR and parsing to precision, review thresholds and per-page pricing, each linked to the post where it matters.

How Extend's extraction schemas, test sets and workflows fit together, from its docs, with a worked schema for one made-up invoice.

Extend, Datalab, Google, Azure and Mistral compared on one disclosed invoice task, using only documented features and published prices.

Branches, commits, worktrees and pull requests with a coding agent, plus the git actions an agent should never take without you.

Run Claude Code and Codex non-interactively and in GitHub Actions: flags, the GitHub app, Codex cloud, secrets, and a worked workflow.

How Claude Code and Codex hooks work, the settings.json shape, and a worked hook that runs your checks before the agent commits.

Install Claude Code and Codex, sign in, give each the same broken test to fix in a practice project, and check what it changed before you keep it.