* feat(website): rebuild site on Primer Brand design prototype
Replace the Astro + Starlight site with the Brand Engineering design
prototype, ported component-for-component onto plain Astro + React
islands. The prototype is treated as the authority on markup and
styling; dynamic data is injected into its components rather than the
components being reinterpreted.
Framework:
- Remove @astrojs/starlight entirely, along with its document shell,
search, footer and language selector. BaseLayout.astro now owns the
document head, CSP, social meta and analytics.
- Add @primer/react-brand and @astrojs/react. Alias the package to its
ESM build in vite.resolve, since the default CJS entrypoint breaks
named-export detection during SSR, and mark it noExternal so its
stylesheet imports resolve.
- Promote pagefind to an explicit devDependency; it was previously
pulled in transitively by Starlight.
Pages: home, the five catalogs, the five detail routes, contributors,
Playbook index and articles, and the cookbook are all rendered by
ported prototype components inside a shared PageShell.
Detail pages share a DetailChassis (hero, breadcrumbs, sticky TOC with
scroll-spy, prev/next) while keeping what makes each type distinct: a
file switcher over bundled skill assets, an included-items grid and
external provenance for plugins, and a hero-scale preview for
extensions.
i18n: resolve translated Playbook entries via Astro.currentLocale
inside the shared article route instead of separate [locale] routes.
The explicit routes collided with the i18n fallback routes, so
translated articles were being shadowed by their English originals and
900 nonsensical double-locale pages were emitted.
Search: TopNavSearch now queries the Pagefind index client-side in
addition to the static resource index, merging on href and degrading
to the static index in dev, where no index has been built.
Also delete the superseded vanilla-TS page renderers and Astro
partials, which the ported React components fully replace.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): resolve accessibility violations in ported detail pages
Fixes the three axe violations surfaced by the a11y audit after the
Primer Brand redesign:
- aria-prohibited-attr: `aria-label` was set on roleless `div`s in
InstructionDetail and PluginDetail. The "Applies to" list now uses
list/listitem roles; plugin provenance uses a group role.
- scrollable-region-focusable: the install command `code` element
overflows horizontally but was not keyboard reachable. Added
tabIndex to all three render sites.
- color-contrast: the Playbook "New" label used the brand's
success-fg on success-subtle, reaching only 4.09:1 in light mode.
Stepped one down the same green ramp for 6.14:1. Scoped to light
mode; dark mode already passed and its green-7 is near-black.
Also corrects the stale route list in the audit script: /hooks/,
/workflows/ and /tools/ have never existed as pages.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): collapse long filter facets and unpin detail-page footer
Two issues surfaced by review of the redesigned site:
Catalog facets rendered every option. The prototype's filter groups were
built from small hardcoded arrays, but real data produces 193 tool options
on /agents/ and 245 "Applies to" values on /instructions/. The sidebar grew
to ~10,000px and stretched the whole catalog row, pushing the (already
present) pagination control far below the fold so it read as missing.
Adopt the prototype's own solution for this, which it had already applied to
the extensions page: collapse groups past 10 options behind a "Show N more"
toggle, and cap .filterOptions with an internal scroll area. Ported verbatim
to the agents, instructions, skills, and plugins catalogs.
Detail pages scroll inside .scrollHost rather than the document, but the
footer came from PageShell, outside that element, so it stayed pinned over
the content instead of appearing at the end. The prototype renders its
footer inside the scroll host; PageShell now takes a renderFooter flag so
DetailChassis can do the same. LearningArticleLayout already did this.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): source contributor count from .all-contributorsrc
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Removing playwright-mcp files
* fix(website): resolve article text colour inside themed scope
Injected markdown inherited color from body, which sits outside the
ThemeProvider and always resolved the light-mode token, making body
copy unreadable in dark mode.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): drop 'In this article' TOC from resource detail pages
Resource detail pages (agent, instruction, skill, plugin, extension) are
not articles, and their markdown headings do not form a meaningful
outline. Playbook and cookbook articles keep their TOC.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* style(website): thin the sidebar scrollbar on detail pages
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): use themed link colour for markdown links in dark mode
Raw markdown-injected <a> elements previously fell back to the browser's
default blue/purple link colours, which are harsh against the dark-mode
background. Route them through --brand-color-text-link-rest (and the
pressed/hover token) instead.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): apply ThemeProvider to Playbook article layout so dark mode works
LearningArticleLayout replaces PageShell for Playbook articles but never
wrapped itself in a ThemeProvider, so its useTheme() call always fell back
to Primer's light default regardless of the site's actual theme preference.
Split the component into a thin ThemeProvider wrapper plus the existing
implementation (now LearningArticleLayoutBody), matching the pattern already
used by PageShell.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): align raw markdown code blocks with prototype styling
Astro's default Shiki config bakes in the fixed 'github-dark' theme's
literal colors, ignoring the site's actual light/dark mode - this made
plain markdown-fenced code blocks (as opposed to the prototype's own
SyntaxHighlightedCode component) always render a hardcoded dark box
regardless of theme.
Switch shikiConfig to the 'css-variables' theme so highlighted tokens
resolve through --astro-code-* custom properties instead, then map
those to the same brand color tokens the prototype's codeBlock uses
(canvas-subtle background, border-muted border, brand text/link/accent
colors for tokens). Raw markdown code blocks now match the prototype's
bordered, canvas-subtle surface in both color modes.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): show real contributor count on every page shell
The contributor badge rendered 0 on Playbook, Cookbook, home and custom
pages, and reverted to 0 on hydration everywhere else.
Two causes:
- Shells that bypass PageShell (LearningArticleLayout, PlaybookIndex,
PlaybookArticleBody, CookbookIndex, HomePage, TopNav, Custom) defaulted
contributorsTotal to 0 instead of the site-data value.
- site-data read .all-contributorsrc with node:fs at module scope. Those
shells are client:load hydrated, so the read threw in the browser and
the count reset to 0 after hydration.
The count is now read once in astro.config.mjs and inlined through
vite.define as __CONTRIBUTORS_TOTAL__, so it is a literal in both the
server render and the client bundle.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* style(website): soften catalog filter list scrollbars
The filter option lists only set scrollbar-width: thin, so they rendered
the platform default scrollbar. They now use the same muted, transparent
track treatment as the article sidebar, which resolves through
--brand-color-border-muted in both colour modes.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* refactor(website): remove Access from tools panel from plugins catalog
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* feat(website): make Copilot app deep link the default plugin install
Plugin detail pages exposed only a copyable CLI command. They now lead
with a ghapp://plugins/install deep link in the same split-button
ActionMenu the other detail pages use, keeping the CLI command available
as a Copy action in the menu.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): migrate markdown remark plugins to unified processor
Astro 7 deprecates markdown.remarkPlugins in favour of passing a
unified() processor from @astrojs/markdown-remark. Moves the GitHub
admonitions plugin into markdown.processor, clearing the startup
deprecation warning. shikiConfig stays at the markdown level as it is
not part of UnifiedProcessorOptions.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): fail the build when the contributor manifest is unreadable
Silently falling back to 0 is how the contributor badge regressed before,
so a missing or malformed .all-contributorsrc now throws in production
builds and warns in dev instead of shipping a wrong count.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Fix codespell and CodeQL findings in website components
- Fix real typos flagged by codespell: 'Couldn't' -> 'Couldn't'
(plain apostrophe, matching convention elsewhere in JSX) and
'Unparseable' -> 'Unparsable' in catalogFilters.ts
- DetailChassis.tsx: replace sequential HTML entity unescaping with a
single-pass replace to avoid double-unescape/injection risk flagged
by CodeQL
- SyntaxHighlightedCode.tsx: make the markup HTML comment regex match
newlines so multi-line comments cannot break out of the token
(Bad HTML filtering regexp)
- pagefindSearch.ts: strip HTML tags in a loop until stable so nested/
malformed markup can't survive a single-pass strip (Incomplete
multi-character sanitization)
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): harden detail heading sanitization
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* github-app settings
* Fix light-theme flash before dark mode applies on page load
ThemeProvider colorMode="auto" from @primer/react-brand only resolves
the real OS colour-scheme preference inside a useEffect, so its first
render is always light. Add a synchronous inline script in <head> that
reads prefers-color-scheme and stamps data-color-mode onto <html>
before first paint, using the same attribute Primer Brand's CSS
already keys off. React's own data-color-mode on the inner element
takes over once it hydrates.
Fixes#2820
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Fix light-mode scrollbar rendering in dark mode on Playbook articles
.scrollHost (the article scroll region shared by Playbook articles,
cookbook recipes, and extension detail pages) never declared a
color-scheme, so the browser always painted its native scrollbar using
light-mode chrome regardless of the site's active theme. Set
color-scheme: light dark as a baseline and pin it explicitly to the
resolved data-mode, and add scrollbar-color so Firefox picks up the
themed thumb colour too.
Fixes#2822
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Widen and center extension card preview image
The Canvas Extensions overview card thumbnail was fixed at 360px, leaving
uneven left/right margins within the card. Let it grow up to 440px and
center it with margin-inline: auto so the image is balanced within the
card.
Fixes#2823
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Link Submit CTAs to each resource's contribution section
The bottom "Submit" CTA on every catalog page pointed at the top of
CONTRIBUTING.md, forcing readers to scroll and hunt for the relevant
"how to contribute" instructions for that specific resource type.
Point each catalog's CTA directly at the section that documents how to
contribute that resource: agents, instructions, plugins, and skills
link to their dedicated docs/README.*.md "How to Contribute" section,
and canvas extensions link to CONTRIBUTING.md's "Adding Canvas
Extensions" section (extensions have no dedicated README doc).
Fixes#2824
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Center-align "View on GitHub" source link with its icon
The source-link label sat on the default inline text baseline instead
of being vertically centered against the GitHub mark icon next to it,
so the two visually drifted apart. Wrap the icon and label in an
inline-flex container with align-items: center so they share the same
vertical center.
Fixes#2825
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Fix mobile resource and Playbook article layout issues
Clean up the shared resource detail and Playbook article styling so mobile pages no longer show unintended divider lines or overflow horizontally.
- remove article-section divider borders so content separates with spacing instead of white/muted horizontal rules
- drop the detail-page breadcrumb divider on mobile to match the Playbook mobile treatment
- constrain raw markdown pre/code blocks to their column and let long code scroll horizontally
- allow metadata chips to wrap inside the mobile sidebar instead of widening the page
Fixes#2827Fixes#2828Fixes#2829
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Only show language selector on pages with real translations
Most of the site is English-only, but the language selector rendered
unconditionally on every page even though only the
learning-hub/copilot-workshops/app track has mirrored translations.
Add hasTranslations() to playbook-routes.ts, thread a showLanguageSelect
prop through PageShell, TopNav and LearningArticleLayout, and compute
it from the article's englishId in learning-hub/[...slug].astro so only
translated Playbook articles show the selector.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Add Workshop recommended card to Playbook index
The copilot-workshops hands-on tracks (VS Code, CLI, App, Cloud harnesses) existed and rendered correctly at direct URLs, but had no entry point in the new Playbook index UI -- the Articles grid intentionally excludes nested multi-page tracks by design, and the only prior link was buried in body prose.
Adds a 'Workshop' recommended card (following the existing cli-for-beginners precedent) linking to /learning-hub/copilot-workshops/, and registers the route in pageHref.ts.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Add a Workshop content-type filter to the Playbook
The hands-on workshop tracks (CLI for Beginners, and the copilot-workshops harnesses) had no shared way to discover them from the Articles grid -- each had a recommended card, but no consistent categorization.
Tags both workshop landing pages 'workshop', adds a new 'Workshop' Kind/content-type facet (derived the same way as the existing Terminology/Tutorial/Example kinds), and includes both landing pages in the Articles grid dataset so filtering by Content type > Workshop surfaces exactly the two tracks.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Fix P0/P1 items from Brand Engineering audit (brand-experience#458)
- P0: html lang now reflects the actually-rendered locale for a page,
not the requested URL locale (BaseLayout.astro, playbook-routes.ts,
PlaybookArticle.astro)
- P0: markdown tables get tabindex=0 via enhanceMarkdownA11y, which
Playbook articles were never running (PlaybookArticle.astro)
- P0: long inline content and breadcrumbs no longer clip on narrow
viewports (dotnet-upgrade.module.css, github-copilot-app.module.css)
- P1: strip the markdown document's own leading H1 so detail pages
don't render the title twice (detail-page.ts)
- P1: document the hero secondary-CTA hex and InstructionsCatalog
illustration colors as intentional prototype-fidelity values rather
than defects (styles.module.css)
- P1: add explicit Heading size props in PlaybookIndex.tsx
Ref: github/brand-experience#458
* Rename Playbook to Learning Hub across UI and code
Renames PlaybookIndex.tsx -> LearningHubIndex.tsx, PlaybookArticle.astro
-> LearningHubArticle.astro, PlaybookArticleBody.tsx ->
LearningHubArticleBody.tsx, PlaybookIcon.tsx -> LearningHubIcon.tsx,
lib/playbook-routes.ts -> lib/learning-hub-routes.ts, and
lib/playbook-article.ts -> lib/learning-hub-article.ts. Updates all
call sites, UI copy, TopNav's playbookLabel prop, nav/search entries,
and home page CTA data to use Learning Hub terminology. URLs under
/learning-hub/* are unchanged.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Port github-copilot-app as bespoke Learning Hub article
Adopts the prototype's github-copilot-app.tsx directly as a dedicated React
component instead of rendering it through the generic markdown pipeline,
preserving its video carousel, comparison list, learn-more band, and other
bespoke JSX exactly as designed.
- New GithubCopilotApp.tsx under components/brand/learning-hub/, ported
near-verbatim from the prototype (data-wiring only: pageHref import swap,
media paths).
- The two large source videos (~23MB, ~32MB) are not committed; the article
falls back to their poster/webp images with a TODO to host them externally.
- learning-hub-bespoke-articles.ts lists slugs that bypass the generic
LearningHubArticle.astro pipeline.
- [...slug].astro branches to the bespoke component for "github-copilot-app"
(English only for now); all other slugs and locales keep the generic path.
* Port bespoke Learning Hub articles
Add bespoke Learning Hub article components for the remaining prototype-backed
article pages, wire the cookbook index to the data-driven port, and route the
CLI overview page through the new bespoke component.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Resolve merge validation blockers in astro config
* Regenerate README docs after merging origin/main
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Fix Learning Hub Get started pages disappearing after opening (#2960)
The bespoke Learning Hub article components (GithubCopilotApp,
AgentsAndSubagents, CopilotConfigurationBasics,
GithubCopilotTerminologyGlossary, UsingAutomationsInCopilotApp,
WhatAreAgentsSkillsInstructions, WorkingWithCanvasExtensions,
CliForBeginnersOverview, CookbookIndex) received the `pageHref` helper
as a function prop passed into a `client:load` React island. Functions
cannot survive Astro's client-hydration prop serialization, so the
prop arrived as `undefined` on the client. The page rendered correctly
server-side (visible briefly), then React threw
`TypeError: pageHref is not a function` during hydration and unmounted
the tree, making the page appear to disappear.
Fixed by having each component import the `pageHref` singleton
directly (as AgentDetail, HomePage, and the other catalog components
already do) instead of receiving it as a prop, and removed the now
unnecessary `pageHref={pageHref}` prop and its import from the two
Astro routes that render these islands.
Verified in the dev server: all 9 previously affected routes now
hydrate without error, and `npm run build` completes successfully
(1097 pages).
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* Strip stray trailing divider from Learning Hub articles (#2961)
Several Learning Hub markdown articles end their body with a trailing
--- thematic break, left over from an authoring template. Rendered
as-is it produced an unlabeled <hr> sitting directly above the page
footer, with no heading/section boundary to justify it.
Fix this generically in buildArticleSections() by stripping a single
trailing <hr> from the rendered HTML before it's split into sections,
rather than hand-editing every affected content file. This also
protects future articles authored with the same trailing-rule habit.
Legitimate mid-content dividers (e.g. between CLI lesson steps) are
left untouched.
* Remove emojis from Learning Hub article nav (#2962)
CLI for Beginners lessons decorate their ## headings with emoji
(e.g. "🎯 Learning Objectives"), which is fine inline in the article
body but read as stray glyphs once surfaced as plain-text labels in
the "In this article" nav — other Learning Hub pages don't
emoji-decorate their nav, so this was inconsistent.
Add a shared stripEmoji() helper and apply it to the nav labels built
from markdown headings in LearningHubArticle.astro. The article body
headings themselves are untouched.
* Fix hero code blocks blending into background (#2963)
Detail page hero install-command code blocks (SyntaxHighlightedCode via
heroExtras) used the same background token as the hero surface itself
(--brand-color-canvas-subtle), making them visually indistinguishable
except for a thin border.
Add a scoped .heroContent .codeBlock override in both shared style
modules (dotnet-upgrade.module.css for catalog Detail pages/Cookbook,
github-copilot-app.module.css for Learning Hub) that sets the
background to --brand-color-canvas-default instead, so hero code
blocks stand apart from the hero background.
The base .codeBlock rule is left untouched since body-content code
blocks (e.g. Cookbook recipe steps) correctly rely on canvas-subtle
being distinct from the page's canvas-default background.
* Tie color-scheme to dark mode for native scrollbars (#2964)
Native scrollbars (the page scrollbar, and any overflow container
without custom scrollbar styling, e.g. code block <pre> elements) are
painted by the browser based on the color-scheme CSS property, not our
design tokens. Only a handful of scoped containers (.scrollHost,
.filterOptions, .sidebarSticky) had custom scrollbar-color rules; the
page scrollbar and plain code-block overflow scrollbars had no
color-scheme set at all, so they rendered with light OS-default
scrollbars even when data-color-mode="dark" was set on <html>.
Add a global rule tying html's color-scheme to the same
data-color-mode attribute BaseLayout.astro already stamps on <html>,
so every native scrollbar (and other UA-drawn form controls) follows
the site's theme.
* Unify article/detail page typography, tables, and dividers (#2965)
Raw markdown injected into .articleSection (Skill/Agent/Instruction/
Plugin/Extension Detail pages via dotnet-upgrade.module.css, and
Learning Hub articles via github-copilot-app.module.css) had no styling
for headings, tables, horizontal rules, blockquotes, or inline code, so
it fell back to inconsistent browser UA defaults -- most visibly,
tables rendered with no borders or header emphasis at all.
Add a shared set of rules to both CSS modules, scoped under
.articleSection, that:
- map raw h1-h6 onto the brand type scale (font, weight, line-height)
- give hr a full-width divider matching the border-muted token
- style blockquote with a left accent bar
- render inline `code` (outside <pre>) as a small chip, matching
.inlineCode used elsewhere for hand-authored tokens
- style table/th/td with bordered cells and a canvas-subtle header row,
scrolling horizontally on narrow viewports
Also fix skills/acquire-codebase-knowledge/SKILL.md: a stray blank
line inside the "Bundled Assets" table split it into a lone header row
and a second block with no header, so GFM table parsing stopped after
the first row and the remaining rows rendered as literal pipe-delimited
text instead of a table -- this was the exact bug shown in the issue
screenshot.
* fix: align previous/next navigation links horizontally on Detail pages
Detail page Previous/Up next links were stacked vertically. Change
.nextUp to a row layout (space-between) so Previous sits left-aligned
and Up next right-aligned on the same row, with a max-width: 40rem
media query reverting to a stacked column layout on small screens.
Use an explicit .nextUpNext modifier class (applied to the "Up next"
link specifically) with margin-inline-start: auto instead of a
positional :last-child selector, so the link right-aligns correctly
even when it is the only link present (e.g. the first item in a
catalog, which has no Previous link).
Fixes#2966
* fix: align callout blocks with article content width
The .proTip callout (Note/Tip/Caution admonitions in Learning Hub
articles, and the "Maintained outside this repository" notice on
external Plugin Detail pages) had its own 32px horizontal margin.
The prototype places this callout as a sibling of .articleSection
(which has no horizontal padding of its own), using that margin to
align its edges with the section's own 32px padding. Our port always
renders it nested *inside* an already-padded .articleSection, so the
extra margin doubled up with that padding, making the callout visibly
narrower than the surrounding paragraph/table text -- most obvious on
narrow viewports.
Remove the horizontal margin (keep only the bottom spacing) in both
dotnet-upgrade.module.css and github-copilot-app.module.css so the
callout's internal padding lines up with the article text on both
edges, at all viewport widths.
Fixes#2967
* fix: standardize mobile card padding across resource pages
Agents, Instructions, Skills, and Plugins catalog cards (.item) used
the desktop 60px padding at every viewport, including phone widths,
because their @media (max-width: 47.99rem) block never reduced it --
unlike the Extension catalog, which drops to a 32px inset at that
breakpoint. This made cards on those four catalogs look inconsistently
over-indented on mobile compared to Extension cards.
Add the same `.item { padding: var(--base-size-32); }` override to the
mobile media query in agents.module.css, instructions.module.css, and
skills.module.css. plugins.module.css already had an override, but
with an asymmetric 32px/24px padding -- normalized it to the same 32px
on all sides used everywhere else.
Fixes#2968
* fix(website): address PR review feedback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): preserve legacy resource links
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): bump transitive js-yaml to patched versions
Force astro's js-yaml dependency to 4.3.2 and front-matter/gray-matter's to 3.15.2 via npm overrides, fixing GHSA-52cp-r559-cp3m and GHSA-5p4m-2wfm-xmqj (quadratic-complexity DoS via YAML merge keys and !!omap resolution). Addresses github/vuln-mgmt#209786.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): address redesign follow-up issues
Fix light-mode favicon contrast with an adaptive SVG mark, allow YouTube embeds through the site CSP, and align resource catalog sort controls with the Playbook article sort treatment across Agents, Instructions, Skills, Plugins, and Extensions.
Addresses #3006, #3007, #3008.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): address PR review feedback
Respect reduced-motion preferences in detail TOC scrolling, restore default search indexes for bespoke Playbook pages, add pagination scroll/focus handling to the Agents catalog, sanitize contributor and cookbook URLs before rendering external links, index extension records in Pagefind, and restore install affordances for marketplace-backed external extensions.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): revert Learning Hub naming from Playbook rename
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* chore(website): update stale Playbook comments to Learning Hub
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): align source label in detail sidebar
Fixes#3207
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
* fix(website): address security review feedback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 80686fef-efe3-4cdd-8cd6-bfa61a5d0af6
Ports four corrections these two skills received upstream after a second trial
run against a real repository. #2951 merged the snapshot taken before them. The
bundled scripts are already identical to their upstream versions and are not
touched here.
1. Neither skill told the agent that text read out of the audited repository is
data rather than instruction. These skills exist to read untrusted
repositories, so a README, a code comment, a commit message or a dependency
manifest reached the model with no framing -- and a line claiming a file is
approved, or telling the audit to skip a module, reads exactly like a
guardrail. Both skills now carry the rule and report such text as a finding
instead of following it.
2. The citation rule allowed anchors to land beside the symbol rather than on
it: the blank line above a definition, a decorator, or a line inside a
multi-line literal. In one trialled file every anchor sat two lines above the
def it named. The rule is now a single applicable test -- the line you cite
must literally contain the thing you name, and a cited range must contain it
on the first line. Quoted text is cited at the line the quoted characters are
on, because a comment has its own line number and it is usually not the line
of the code beside it.
3. "Never restate a count without the raw output in front of you" was ignored
twice in that trial, so the rule flips from prohibition to requirement: any
number stated must appear under Checks Run next to the command that produced
it. Unwilling to show the command means describing the pattern rather than
counting it.
4. Both Related Skills sections said the skill is one of seven and that the
other five cover the remaining ground. Six, not five. Each section now names
its sibling in this repository and links the remaining five out.
Front matter is unchanged, so the generated README tables do not move.
* feat(skills): add test-gap-audit and docs-sync-audit
Two repo-agnostic review skills that answer questions the existing testing and
documentation skills do not.
test-gap-audit asks which behaviour is not covered, rather than how to write a
test in a given framework. Given no scope it audits the whole repository,
inventories the testable surfaces, and reports which routes, services, jobs and
contracts have no tests, too few assertions, or only indirect coverage. It
bundles coverage_map.py, which detects the test framework and naming convention,
then matches every source file to tests by name, by mirrored path, and by what
the test files actually import, and ranks the unmatched by risk keyword and size.
docs-sync-audit compares what the docs claim against what the code does. It
bundles docs_drift.py, which checks documented npm scripts and make targets
against the ones that exist, relative Markdown links against the filesystem, and
environment variable names in both directions. It also reports a documented
setting that is read only inside a module nothing imports, which is configuration
that reads as working but cannot take effect.
Both are read-only: they report and do not edit unless asked. Both emit the same
contract, so a finding always carries a P0-P3 severity and a path:line you can
open. Both scripts are Python standard library only, install nothing, and are
accelerators rather than requirements, so each skill still works when the script
cannot run.
The existing testing and docs skills here are framework-specific, which is where
most of the value is. These are the repo-agnostic complement: pytest-coverage
raises coverage inside a pytest project, and this decides where coverage is
missing across a repository regardless of language.
* fix: satisfy codespell and regenerate all generated docs
Two CI failures on the first push.
codespell flagged `testng` and `shouldBe` in coverage_map.py. Both are
legitimate identifiers rather than typos: TestNG is the Java test framework the
script detects by name, and shouldBe is the Kotlin and Scala assertion method
matched by its assertion-detection regex. Added both to ignore-words-list with a
comment each, following the convention already used for the other entries.
validate-readme failed because I had reverted docs/README.agents.md. `npm start`
rewrites a Dynatrace MCP URL there from re-fetched external plugin data, which is
unrelated to these skills, so I had excluded it to keep the diff scoped. That was
wrong: the check regenerates every generated file and compares, so the commit has
to carry whatever the build produces. Restored.
---------
Co-authored-by: Aaron Powell <me@aaron-powell.com>
trigger_live_flow now runs scheduled (Recurrence) flows on demand;
list_live_connections requires environmentName and applies top after search;
the ChatGPT/claude.ai connector path is documented in the mcp skill; the
brand reads FlowStudio (one word) across the skills and plugin README; the
build skill is back under the 500-line lint limit.
Co-authored-by: Aaron Powell <me@aaron-powell.com>
`-f labels[]="bug"` fails under zsh, the default shell on macOS, because `[]`
is a glob pattern. The argument never reaches gh. Quote the whole
`name[]=value` pair instead.
The skill also states the MCP server cannot create, update, or comment on
issues. github-mcp-server ships issue_write, add_issue_comment, and
sub_issue_write. Documented them and reworded the CLI section; the gh api
examples are unchanged.
* feat: update Apple App Store reviewer guidance
* refactor: narrow App Store guideline checks
---------
Co-authored-by: Jimmy Stridh <jimmystridh@users.noreply.github.com>
FlowStudio MCP 1.2.42 delivers trigger inputs to Button, PowerApps, and Skills
triggers; the skills still said trigger_live_flow was HTTP-only and told agents
to build an HTTP twin. Verified live against server 1.2.43.
Adds a skill that converts text, markdown, or the output of another
skill into an MP3 using Kyutai pocket-tts, a small neural TTS model
that runs on CPU.
No existing skill in the collection generates audio, so this fills a
gap rather than duplicating one. It is designed as a terminal step in
a chain: roundup, daily-prep, or meeting-minutes produce the text,
speak-summary makes it listenable.
Two details worth calling out:
- The bulk of SKILL.md is guidance on rewriting written prose for the
ear before synthesising. Feeding markdown straight into a TTS engine
produces something technically correct and unlistenable, so that
step carries most of the value.
- Synthesis is local and CPU-only, so nothing is sent to a cloud
speech service and the skill works unattended in a headless
container as well as on a laptop.
The bundled script bootstraps pocket-tts into a cached virtualenv on
first use, selecting a Python in the supported >=3.10,<3.15 range
rather than assuming python3 qualifies, and failing with actionable
guidance when none is available.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 1a70aa08-b622-4825-ad63-5a12370add1f
* Add poka-yoke skill: make misuse unrepresentable rather than warned against
Mistake-proofing, applied to code. The skill pushes toward devices that make
a wrong action impossible or self-announcing, rather than instructions asking
someone to avoid it, and refuses to accept a comment or a convention as a fix.
The gap it addresses. Given a design, models readily list what to fix and
rarely state what the fix makes impossible. Measured across 591 blind-graded
runs and six model families, responses named the foreclosed set 42% of the
time unprompted and 81% with this skill applied. Assertions were written
before the runs and graded blind to configuration.
Stated with its limits, because they matter: every run was the first turn of
a fresh session, so this measures the ceiling rather than what survives a
long working session; the comparison was against no methodology rather than a
different one, so it does not establish that this particular method caused
the gain; and the skill costs something measurable, making responses somewhat
worse at spotting the specific defect already on the page while better at
changing the shape that allowed it. All of that is in the skill body rather
than omitted.
Bundled, all self-contained, no network access and no dependencies:
scripts/detect_hazards.py standard-library scanner, 42 pattern rules
across 20 hazard shapes, five languages
references/hazard-catalog.md the taxonomy with device per shape
references/lang-*.md Python, TypeScript, Go and Rust patterns
Verified with npm run skill:validate and npm run build. Raw benchmark runs,
the harness and the assertion checklists are public at
https://github.com/rainmanjam/poka-yoke
* Rename HasTable to HasFrom in the TypeScript typestate example
codespell failed the PR: it reads HasTable as a misspelling of hashtable.
The identifier is a legitimate generic parameter on a typestate builder,
QueryBuilder<HasTable, HasWhere>, so this is a false positive, but it is our
file breaking their gate and the fix costs nothing.
HasFrom is also the better name: the flag tracks whether .from() has been
called, not whether a table exists.
Renamed in the upstream repository too, so the two copies do not diverge on
day one. Verified the whole bundle is codespell-clean locally.
* Correct the effect figures in the poka-yoke skill
The submitted numbers (42% -> 81%) could not be reproduced from the upstream
gradings. Recomputed over the six models the sentence describes: 45% -> 80%
across 132 verdicts. Adds the per-scenario breakdown, which is the more useful
claim: the gains are in tasks where nobody asked for a design review.
* Add repo-standardizer skill: polish any GitHub repo surface
* fix(skill): remove non-string metadata and fix template README refs for vally lint
* docs(skill): drop markdown relative links from README language-switcher example
Vally valid-refs flagged ./README.md and ./README.zh.md as missing file
references; they are generated artifacts, not skill files. Keep the
example as plain text to satisfy the linter.
* fix(template): add missing language switcher to English README template
Align with README.zh.md template and SKILL.md switcher convention:
current language as plain text, other languages as relative links.
---------
Co-authored-by: programmingWTF <programmingWTF@users.noreply.github.com>
The MCP Apps sample HTML bundle imported ext-apps from esm.sh using an
unpinned major-version tag (@1), which Snyk flags as an unverifiable
external dependency / third-party content exposure risk (indirect
prompt-injection surface, since the imported module can call
app.updateModelContext and app.callServerTool). Pin it to the current
release (1.7.5).
Also pin the sample `dotnet add package` commands for the
ModelContextProtocol* and Microsoft.Extensions.Hosting packages to
their current exact versions (2.2.0 / 10.0.11), and correct a stale
"2.0.0 is current" note in packages.md.
* Bump gem-team plugin version to 1.102.0 and add entry
Updated marketplace.json and plugin.json to version 1.102.0, added gem-team plugin entry with 16 items, refined argument hints in multiple agent definitions, and updated planner, reviewer, and skill creator configurations.
* Bump gem-team plugin version to 1.103.0 and update acceptance_criteria references
* refactor: simplify agent definitions and add gem-design-md-guidelines and gem-devops-guidelines skills
* feat: optimize orchstrator for faster workflows
* chore: optimize skills
* chore: update readme
* chore: bump version to 1.108.0
* chore: streamline planner
* Enhance Oracle-to-PostgreSQL migration skills and documentation
- Update migration agent guidelines to prioritize extension tool usage for code migration.
- Refine migration phases with detailed steps for pre-migration review and schema migration.
- Add new reviewing skill references for PostgreSQL materialized view refresh and UNION ALL planner risks.
- Ensure consistency in collation handling and testing strategies across skills.
* - Review migration phases to ensure correct order of execution
- Add exit criteria to each migration phase
- Remove invocation of `ms-ossdata.vscode-pgsql` extension due to dependency on VSCode
- Update README.md to reflect changes in migration phases and exit criteria
- Update broken reference to custom agent in plugin.json
* Enhance downstream migration skills and following custom agent improvements
- Added new skill for migrating .NET data access code from Oracle to PostgreSQL.
- Updated README to include new migration skill.
- Improved existing skills with clearer file naming conventions and migration actions.
- Added reference documents for handling Oracle-specific functions and pagination.
- Created detailed guides for NVL, DECODE, ROWNUM, SYSDATE, and DUAL replacements.
* Clarify PostgreSQL schema immutability and stored procedure migration risks
* Update target schema path in migration documentation for clarity
* fix(skills): clarify Phase 3-only scope for Oracle test skills
Both Oracle test skills were ambiguously worded in ways that could
cause a model to invoke them during Phase 6 (PostgreSQL test migration)
instead of using them exclusively in Phase 3.
Key changes:
- Rewrite descriptions to explicitly state Phase 3-only usage and
warn against invoking during Phase 6
- Replace "scaffold for Oracle first" with "Oracle only" to remove
the false implication of a second PostgreSQL scaffolding step
- Replace "Tests validate behavior consistency when running against
Oracle or PostgreSQL" with clear Oracle-only framing
- Rename "DB-agnostic assertions" → "Assertion portability" and
explain the why (survive Phase 6 migration without rewrites)
- Fix datetime bullet in integration tests skill to use generic
Oracle column precision language instead of PostgreSQL type syntax
- Name Oracle NuGet package explicitly (Oracle.ManagedDataAccess.Core)
* feat(oracle-to-postgres): gate Phase 1 on DDL presence; add DDL scan to Phase 2 risk analysis
- Phase 1 success criteria now requires Oracle DDL artifacts to be
confirmed present at the recorded location before proceeding.
If missing, the agent stops and prompts the user to provide them.
- Phase 2 risk analysis now explicitly scans DDL/Oracle/{ProjectName}/
as supplemental context, summarising procedure complexity indicators
(dynamic SQL, DBMS_* / UTL_* references, autonomous transactions,
pipelined functions, BULK COLLECT/FORALL, REF CURSOR, TYPE bodies)
rather than ingesting DDL files wholesale. This ensures schema-level
migration risk is captured even when it isn't visible in application
code alone.
* - Merge in latest 'main' changes
- Update custom agent plugin (eg resolve conflict and add new skill)
- Validate skills
- Run build
* chore(plugin.json): update version to 1.1.0
* feat(oracle-to-postgres): update version to 1.1.0 for migration expert plugin
* fix: add INOUT to ignore-words-list for PostgreSQL migration
---------
Co-authored-by: TCPrimedPaul <paul.delannoy@tc.gc.ca>
* Add landing-page-conversion-audit and server-side-conversion-tracking skills
* Drop references to skills not present in this repo
---------
Co-authored-by: autonnel <283055447+autonnel@users.noreply.github.com>
While implementing the Agent Plugin v1 spec in https://github.com/stbenjam/skillsaw/pull/483, I used this repository as a canonical real-world source for tests. Along the way its checks surfaced genuine correctness issues worth fixing regardless of tooling:
- Internal links that do not resolve (22): wrong relative paths, links whose targets are runtime output paths rather than bundled files (now inline code), a schemeless external URL parsed as a relative path, and links to files that do not exist (removed).
- Bundled skill files never referenced from their SKILL.md (14): each is now cited from the section where it is used, so nothing ships unreachable from the skill's instructions.
- Unclosed or mis-nested code fences (11): markdown templates embedding code blocks now use four-backtick outer fences so inner fences nest instead of closing the template early and leaking example content as live markdown; also removed stray orphan fences and closed an unterminated block at end of file.
- Invisible unicode (1): removed a zero-width space from a code example, a copy-paste artifact invisible to reviewers.
Adopt the Agent Plugins v1.0.0 namespaced composition model from #2546. Bundle the reusable daily-focus-board canvas into Ember only and remove its obsolete standalone catalog manifest.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: 52be9c67-3ae4-4610-93d0-fe0b7ab95ccb
* Add Azure Developer CLI skill and related documentation
* Add Azure Developer CLI skill to README with usage instructions and bundled assets
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* Update dotnet-mcp-builder skill to ModelContextProtocol 2.x
Align the skill with the C# SDK 2.0.0 release and the MCP 2026-07-28
spec: stable line is now 2.x, HttpServerTransportOptions.Stateless
defaults to true, roots/sampling/MCP-channel logging are [Obsolete]
(MCP9005) with the multi-round-trip input_required pattern as the
replacement, discovery-first negotiation (server/discover) supersedes
the initialize handshake, Mcp-Method/Mcp-Name routable headers, raw
structuredContent for non-object results, required Tool.inputSchema,
and the new ModelContextProtocol.Extensions.Tasks and
ModelContextProtocol.Extensions.Apps packages (typed MCP Apps support
replacing the hand-rolled _meta/ui:// pattern on 1.x).
* Address Copilot review: Apps extension accuracy, header scope, capability ownership
- packages.md: the Apps package replaces the manual _meta wiring, not the
ui:// resource; note the experimental MCPEXP003 diagnostic; label the
1.x -> 2.0 list as highlights and add the OAuth/SSE runtime changes
with a pointer to the full release notes.
- transport-http.md: Mcp-Method is on every POST, Mcp-Name only on named
invocations (tools/call, prompts/get, resources/read) - do not require
it globally at gateways.
- mcp-apps.md: current MIME type is text/html;profile=mcp-app (skybridge
is a legacy draft value); document [McpAppUi] + WithMcpApps().
- server-features.md: roots/sampling are client capabilities, only
logging sits on ServerCapabilities.
* Correct stateful HTTP guidance: 2026-07-28 has no HTTP sessions
Per the official SDK v2 elicitation docs, a server with Stateless=false
refuses the 2026-07-28 revision so dual-path clients fall back to an
initialize-capable revision; ElicitAsync cannot be used on 2026-07-28
Streamable HTTP at all. Reframe stateful HTTP as down-level
compatibility mode and document the multi-round-trip pattern
(InputRequiredException / InputRequest.ForElicitation, retry with
InputResponses -> ElicitResult) as the current-protocol way to ask
mid-tool, across SKILL.md, transport-http.md, and elicitation.md.
* Update typescript-mcp-server-generator skill to MCP TypeScript SDK v2
Replace the retired monolithic @modelcontextprotocol/sdk with the v2
focused packages (server, node, core, framework adapters), require
zod@^4.2 and Node 20+, document the registerTool config-object API,
the ctx handler context, the new error hierarchy, removed SSE/WebSocket
transports, and the v1-to-v2 codemod migration path.
* Address Copilot review: adapter peer frameworks and sampling consistency
Framework adapters now note their required peer framework install
(e.g. @modelcontextprotocol/express + express), and the two remaining
sampling recommendations are replaced with the multi-round
input_required pattern that v2 recommends over the deprecated
sampling subsystem.
Resolve the .codespellrc conflict by retaining both the daily-focus-board checkin key and upstream ACI term.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: fd1eae93-cc9f-4777-812c-a2a9872e1c2b