Repository navigation
docs: add the fighting AI training bias agent skills post - #476
Conversation
Flagship post for the blog-revival series (#379, tracker #384): why the project invests in machine-legible surfaces, argued through the one-shot framing. Ships five hand-authored 1200x630 pixel-art SVG covers plus PNG rasters (hero doubles as og:image and the dev.to/Medium cover), syndication on via publish_to_devto.
|
Warning Review limit reached
Next review available in: 29 minutes You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
WalkthroughAdded a blog post that describes bestax’s AI-readiness strategy, including machine-readable documentation, seven agent skills, generated configuration, component catalog validation, npm pointer files, and planned MCP support. ChangesAI-readiness blog
Estimated code review effort: 1 (Trivial) | ~5 minutes Possibly related issues
Possibly related PRs
Suggested labels: Suggested reviewers: 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Preview DeploymentPreview URL: https://374bd66f.bestax.pages.dev |
There was a problem hiding this comment.
Pull request overview
Adds the flagship blog post explaining bestax’s AI-readable documentation, Agent Skills, component catalog, and planned MCP server.
Changes:
- Adds the syndicated blog post with verified examples and links.
- Adds five pixel-art SVG illustrations.
- Adds corresponding PNG rasters for covers and syndication.
Reviewed changes
Copilot reviewed 1 out of 11 changed files in this pull request and generated no comments.
Show a summary per file
| File | Description |
|---|---|
docs/blog/2026-08-05-fighting-ai-training-bias.md |
Adds the flagship AI-readiness post. |
docs/static/img/fighting-ai-training-bias.svg |
Adds the hero artwork master. |
docs/static/img/fighting-ai-training-bias.png |
Adds the rasterized hero image. |
docs/static/img/fighting-ai-training-bias-docs.svg |
Adds documentation-pipeline artwork. |
docs/static/img/fighting-ai-training-bias-docs.png |
Adds its syndication-safe raster. |
docs/static/img/fighting-ai-training-bias-skills.svg |
Adds Agent Skills artwork. |
docs/static/img/fighting-ai-training-bias-skills.png |
Adds its syndication-safe raster. |
docs/static/img/fighting-ai-training-bias-catalog.svg |
Adds component-catalog artwork. |
docs/static/img/fighting-ai-training-bias-catalog.png |
Adds its syndication-safe raster. |
docs/static/img/fighting-ai-training-bias-node-modules.svg |
Adds package-discovery artwork. |
docs/static/img/fighting-ai-training-bias-node-modules.png |
Adds its syndication-safe raster. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (3)
docs/blog/2026-08-05-fighting-ai-training-bias.md (3)
53-63: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winMove installation commands into
PackageManagerTabs.Line [58] uses a plain
bashfence, and Line [63] placespnpm create bestax@latestinline. The documentation rules require installation and run commands inside<PackageManagerTabs>with the canonicalpnpmfenced block and one command per line. Reformat both installation channels with the repository's existingPackageManagerTabspattern.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/blog/2026-08-05-fighting-ai-training-bias.md` around lines 53 - 63, Reformat both installation channels in the blog section using the repository’s existing PackageManagerTabs pattern. Move the skills CLI command and the scaffolder command into canonical pnpm fenced blocks inside PackageManagerTabs, keeping one command per line and preserving the surrounding explanatory text.Source: Coding guidelines
29-31: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winUse root-relative links for same-site artifacts.
Lines [29]-[31] use absolute URLs for internal resources. Use
/llms.txt,/llms-full.txt, and/docs/api/elements/button.mdinstead.Proposed fix
- [llms.txt](https://bestax.io/llms.txt) - [llms-full.txt](https://bestax.io/llms-full.txt) - a **Markdown twin of every page**: append `.md` to any docs URL, like [bestax.io/docs/api/elements/button.md](https://bestax.io/docs/api/elements/button.md) + [llms.txt](/llms.txt) + [llms-full.txt](/llms-full.txt) + a **Markdown twin of every page**: append `.md` to any docs URL, like [bestax.io/docs/api/elements/button.md](/docs/api/elements/button.md)🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/blog/2026-08-05-fighting-ai-training-bias.md` around lines 29 - 31, Update the internal resource links in the documentation list to use root-relative paths: /llms.txt, /llms-full.txt, and /docs/api/elements/button.md, while preserving their link text and surrounding descriptions.Source: Coding guidelines
19-19: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winRemove promotional wording from the post.
flagshipin Line [19] andmost expensivein Line [69] conflict with the blog rule that prohibits salesy hype and superlatives. Use neutral wording such asmain articleandcostly agent failure.Also applies to: 69-69
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/blog/2026-08-05-fighting-ai-training-bias.md` at line 19, Replace the promotional term “flagship” in the article introduction with neutral wording such as “main article,” and update “most expensive” near the referenced later passage to neutral wording such as “costly agent failure.”Source: Coding guidelines
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@docs/blog/2026-08-05-fighting-ai-training-bias.md`:
- Line 73: In the paragraph around “completeness isn't a habit here,” replace
the comma joining the two independent clauses with a sentence boundary, while
preserving the existing wording and meaning.
---
Nitpick comments:
In `@docs/blog/2026-08-05-fighting-ai-training-bias.md`:
- Around line 53-63: Reformat both installation channels in the blog section
using the repository’s existing PackageManagerTabs pattern. Move the skills CLI
command and the scaffolder command into canonical pnpm fenced blocks inside
PackageManagerTabs, keeping one command per line and preserving the surrounding
explanatory text.
- Around line 29-31: Update the internal resource links in the documentation
list to use root-relative paths: /llms.txt, /llms-full.txt, and
/docs/api/elements/button.md, while preserving their link text and surrounding
descriptions.
- Line 19: Replace the promotional term “flagship” in the article introduction
with neutral wording such as “main article,” and update “most expensive” near
the referenced later passage to neutral wording such as “costly agent failure.”
🪄 Autofix
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro Plus
Run ID: 851455c3-ffcb-4a1b-9350-3ddeb324e54f
⛔ Files ignored due to path filters (10)
docs/static/img/fighting-ai-training-bias-catalog.pngis excluded by!**/*.pngdocs/static/img/fighting-ai-training-bias-catalog.svgis excluded by!**/*.svgdocs/static/img/fighting-ai-training-bias-docs.pngis excluded by!**/*.pngdocs/static/img/fighting-ai-training-bias-docs.svgis excluded by!**/*.svgdocs/static/img/fighting-ai-training-bias-node-modules.pngis excluded by!**/*.pngdocs/static/img/fighting-ai-training-bias-node-modules.svgis excluded by!**/*.svgdocs/static/img/fighting-ai-training-bias-skills.pngis excluded by!**/*.pngdocs/static/img/fighting-ai-training-bias-skills.svgis excluded by!**/*.svgdocs/static/img/fighting-ai-training-bias.pngis excluded by!**/*.pngdocs/static/img/fighting-ai-training-bias.svgis excluded by!**/*.svg
📒 Files selected for processing (1)
docs/blog/2026-08-05-fighting-ai-training-bias.md
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
There was a problem hiding this comment.
Deep review — 0 blocking · 2 advisory
| # | Severity | Area | Finding | Location |
|---|---|---|---|---|
| 1 | 🔵 Advisory | Robustness | Deliberate, author-directed deviations from the issue spec (flat .md not folder-style, slug-based canonical_url, publish_to_devto: true with no published: key instead of a draft) — all documented in the PR body and consistent with the three prior syndicated posts |
docs/blog/2026-08-05-fighting-ai-training-bias.md:1 |
| 2 | 🔵 Advisory | Robustness | The in-body hero banner is an SVG markdown image; after rewrite it lands on dev.to as https://bestax.io/img/....svg, and some dev.to/Medium clients render body SVGs inconsistently. Low impact: the actual cover_image is a PNG, the four section images are PNG, and this matches the documented cover convention |
docs/blog/2026-08-05-fighting-ai-training-bias.md:13 |
Overall: This is a clean, unusually well-sourced docs-only PR. I chased every cited capability to its source and all of them hold: the 87-component / 149-line catalog and its CI completeness guard, the seven skill descriptions, the two install channels (npx skills add matches skills/README.md verbatim; create-bestax writes .claude/skills/, .claude/launch.json, and a flavor/prefix/icon-aware CLAUDE.md), the 5.8.0 tarball pointer files (PR #345, merged), the llms.txt/.md-twin pipeline, the AiReady cards with MCP marked coming-soon, and the two verbatim quotes ("Feed it everything..." and "Select an archetype...build it in one shot"). The framing-honesty bar from the issue is met: "bias" appears exactly twice in prose, explicitly flagged as the author's word, not the repo's. The riskiest surface for a docs post with onBrokenLinks: 'throw' is link integrity and syndication, and I verified both empirically rather than by eye.
Residual risk:
- Broken internal links — refuted:
pnpm exec turbo run build --filter=@allxsmith/bestax-docscompleted exit 0 ([SUCCESS] Generated static files), which meansonBrokenLinks: 'throw'validated all 14 internal/docs/and/blog/links, including the date-routed July post URL and the slug-routed v5 post URL. - Syndication drift — refuted: inspected
build/.devto-publish/2026-08-05-fighting-ai-training-bias.md;publish_to_devtois stripped (count 0),cover_imageand every markdown image + internal link are rewritten to absolutehttps://bestax.io/..., and code fences are untouched. - Live-demo props that don't exist — refuted:
Tooltip.label,Badge.dot/color/overlap(overlap="circle"is inBadgeOverlap), andAvatar.nameall exist; the demo mirrors Badge's own JSDoc example. Rasterizer contract also satisfied — all five SVGs declarewidth="1200" height="630"with a full-bleed background rect;pnpm format:checkclean on the post.
🏄 Total glassy conditions on this one, dude — every claim paddled straight back to a real source, the build barreled through clean, and the dev.to rewrite landed the trick. No wipeouts in sight. Ship it and go catch the next set. 🌊
|
On the three nitpicks: PackageManagerTabs: declining. That rule is a docs-tree convention; this post syndicates, and the dev.to preprocessor rewrites markdown only, so a JSX wrapper would land on dev.to as literal Root-relative links for llms.txt, llms-full.txt, and the .md twin: declining. Those are generated build artifacts, not Docusaurus routes, so root-relative markdown links to them trip Promotional wording: split the difference in fefb496. "Flagship" is gone. "The most expensive agent failure" stays: it compares two failure modes (a loud wrong prop vs. a silent rebuilt component), it isn't a superlative about the product, and "costly" would drop the comparison the section is built on. |
Preview DeploymentPreview URL: https://b60e5d02.bestax.pages.dev |
docs: fold the #476 syndication and skill-sync lessons into the CLAUDE.md files
|
🎉 This PR is included in version 5.8.1 🎉 The release is available on: Your semantic-release bot 📦🚀 |
|
🎉 This PR is included in version 4.0.2 🎉 The release is available on: Your semantic-release bot 📦🚀 |
|
🎉 This PR is included in version 1.0.0 🎉 The release is available on: Your semantic-release bot 📦🚀 |
|
🎉 This PR is included in version 2.0.1 🎉 The release is available on: Your semantic-release bot 📦🚀 |
Closes #379. Post six of the blog-revival series (#384), the flagship.
What this adds
docs/blog/2026-08-05-fighting-ai-training-bias.md, the flagship post on why bestax invests in AI-readable surfaces. The body's working vocabulary is one-shot/few-shot and machine-legible; the word "bias" appears in the title and exactly twice in prose (the pivot paragraph, framed explicitly as the author's word, and the closing couplet). The July monoculture post is linked as prior art rather than re-argued.docs/static/img/: five hand-authored 1200x630 pixel-art SVG masters (house style, perrasterize-cover.mjscontract) plus their PNG rasters. The hero doubles asimage/cover_image; the four section images embed as PNGs for syndication safety.Deliberate deviations from the issue spec
.mdinstead of folder style: matches all three previously syndicated posts;devto-preprocessor.jswritesbuild/.devto-publish/<basename>, so a folder post would emit a collidingindex.md, and/img/-rooted markdown images are the proven URL-rewrite path.canonical_urlis slug-based (https://bestax.io/blog/fighting-ai-training-bias), matching the current convention, not the issue's date-path form.publish_to_devto: trueand nopublished:key (the issue said draft): this post syndicates to dev.to and Medium per the author's direction; actual posting stays the manual flow below.Verified
pnpm allgreen; docs build passes withonBrokenLinks: 'throw';pnpm format:checkclean.build/.devto-publish/2026-08-05-fighting-ai-training-bias.mdinspected: images and internal links rewritten to absolute production URLs,cover_imageabsolute, code fences untouched,publish_to_devtostripped.Post-merge (manual)
docs/build/.devto-publish/2026-08-05-fighting-ai-training-bias.mdinto the dev.to editor and publish.https://bestax.io/blog/fighting-ai-training-biasso the canonical source is recorded; the cover PNG is the feature image.Summary by CodeRabbit