v0.3.5

Highlights

• Added new workshop content: Step 29 on teaching agents domain knowledge with Skills, and side quests on skill injection strategies, A/B round-robin experiments, repo poisoning, and the agentic-workflows skill dispatcher.
• Expanded theme-aware (light/dark) diagrams across many steps, including preflight troubleshooting, cache-memory loops, runner label dispatch, audit lifecycle, permission-error decisions, MCP agent loop, and PR reviewer architecture.
• Brand and accessibility pass: migrated multiple workshop SVGs to the GitHub brand palette and fixed several color-contrast violations.
• Consistency fix across dozens of files: standardized Checkpoint headings to use the :white_check_mark: shortcode instead of raw Unicode checkmarks.
• Numerous workshop readability and UX improvements: wider markdown body on large desktops, fixed page gutters and touch target sizing, wrapped long code lines, and clarified gh-aw permission, environment, and init-behavior guidance.
• Ongoing "add gh-aw doc links" passes threading official documentation references throughout the workshop.

Changes since v0.3.4

• Diagram and SVG fixes: architecture clarifications, contrast fixes, brand-palette migrations, and new theme-aware diagrams for steps 07d, 08b, 14, 17, 20–25, and 27 (#2449, #2456, #2458, #2460, #2554, #2559, #2588, #2668, #2711, #2730, #2799, #2826, #2841, #2860, #2874, #2893, #3053, #3145, #3285)
• Checkpoint heading consistency fixes replacing raw Unicode ✅ with the :white_check_mark: shortcode (#2553, #2665, #2716, #2728, #2748, #2763, #2775, #2788, #2797)
• New workshop content: Step 29 on Skills (#2657), skill injection strategies side quest (#2836), A/B round-robin deep-dive (#2867), repo poisoning improvements (#2729, side-quest-17-07 #2827), agentic-workflows skill dispatcher primer (#3080), failure-modes primer extraction (#2949)
• Workshop documentation link passes adding gh-aw doc references across many files (#2468, #2483, #2555, #2669, #2718, #2734, #2802, #2815, #2828, #2875, #2895, #2937, #3068, #3434, #3641, #3689, #3970)
• Guideline violation cleanups across batches of files (#2479, #2538, #2574, #2642, #3677, #3882)
• UX and accessibility fixes: workshop page gutters and touch targets (#2490), wide-code-line wrapping (#2674), wider markdown body on desktop (#2712), image width cap increase (#2721)
• Content clarity and correctness fixes: safe-outputs array syntax (#2489), second-person voice in data-ops checkpoints (#2492), note-style security callout (#2491), permission-error alert downgrade (#2593), broken OpenAI links (#2592), ordered list numbering (#2659), bare code fence language labels (#2658), environment reference clarity (#2840), credential guidance (#7b05e68 / #2868), gh aw init behavior sync (#3111), workflow_dispatch reference link (#6b7ce60 / #4047), gh aw add access model and private:true docs (#4167), research node source URLs (#8435280 / #3177), broken gh-aw overview anchor (#4ec6259 / #3070), Markdown emphasis lint fix (#3985)
• Infrastructure: upgraded agentic workflows with prerelease auto-upgrades enabled and switched to GitHub.com artifact actions (#2935, #2936)
• Golden Ticket and README updates: added welcome callout, removed duplicate callout, added demo environment request notes for facilitators
• Workshop reader typography refresh (#4176)

Generated by 📦 Manual SemVer Release · copilot · auto · 12.6 AIC · ⌖ 5.53 AIC · ⊞ 6.6K