[ntucker]

Requested by Nathaniel · project thread

Motivation

"Debugging with agents" on the Debugging page explained internals to humans, and agents reading the page's .md/llms version got the same human prose plus the manual Redux DevTools walkthrough, instead of the debugging guide the skill gives them.

Solution

• Site (<SiteOnly>): leads with what globalThis.__DC__ gives agents, one example prompt, the browser tools it needs, and an issue link plus a prompt that has the agent draft the bug report (searches for duplicates, follows the bug template, leaves out app data, shows the draft first). Manual debugging is site-only.
• Agents (<AgentOnly>, in .md and llms.txt): the whole docs/core/_agents/devtools-debugging.md guide (the same source as the skill's devtools-debugging reference), plus one line telling the agent to point the user to the browser extension if they want to inspect manually.
• remarkFramework removes <AgentOnly> on the site before the table of contents is built, so agent-only content can hold headings and partials.
• In page markdown, links to _agents pages (browser setup guides) go to the skill's generated copy on GitHub (agentReferenceUrl), since those pages aren't published.

The #debugging-with-agents anchor is unchanged.

🤖 Generated with Claude Code

https://claude.ai/code/session_01GdtHs1YHGyHL3LtFQtD1W7

---

Note

Low Risk
Documentation and static-site markdown generation only; no runtime app or API behavior changes.

Overview
Splits the Debugging page so humans and agents get different content: site readers see a shorter Debugging with agents section (__DC__, example prompts, required browser tools, issue/report prompts) while Manual debugging stays site-only; agent-facing .md / llms.txt output keeps the full devtools-debugging guide via <AgentOnly> plus a line pointing users to the Redux DevTools section on the site.

Pipeline changes: remarkFramework now drops <AgentOnly> when building the Docusaurus site (default) but preserves it when docsToMarkdown runs with keepAgentOnly: true, so agent-only blocks can include headings and _agents/ partials without polluting the site TOC. llms-plugin resolves links to unpublished /_agents/ routes to the matching skill reference on GitHub (agentReferenceUrl). Contributor rules in .claude / .cursor skills-sync docs are updated to match.

^{Reviewed by Cursor Bugbot for commit 94e098a. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Docs preview: https://docs-site-pr-4447-data-client.vercel.app

Built from 94e098a merged into master (this build).

[changeset-bot]

⚠️ No Changeset found

Latest commit: 94e098a

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[changeset-bot]

⚠️ No Changeset found

Latest commit: 4aa1c42

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-23T18:14:43.254111Z	94e098a	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.