Open
Conversation
Complete ground-up redesign: new page structure (/work, /lab, /contact), Geist Sans + JetBrains Mono fonts, dark-mode-only design tokens, case study pages, polished components with gradient placeholders, pill buttons, backdrop-blur nav, contract trace hero animation, and consistent typographic hierarchy across all pages. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…ow hero asset Adds a canvas-based visualization to the hero showing an agent reasoning pipeline (Input → Reason → Memory/Execute/Context → Synthesize → Render/Stream) with animated particles, glowing nodes, bezier curve paths, cycling annotations, and mouse interaction. Includes dark-mode design system, theme toggle, philosophy section, and updated nav/layout. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
… mode system Complete aesthetic overhaul: warm cream light mode with sage green accents, dawn-dark mode with horizon glow. Scroll-driven chat interaction builds up, then tilts into 3D to reveal component layers (reasoning, tools, response, generative UI) with sketched SVG connection lines and sequential highlighting with explanation panels. System preference detection for light/dark with no toggle. Grain overlay, rounded soft UI, all sans-serif typography. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…, and polish - Restore Matter Regular font, monochrome palette, and Rubric Labs nav style - Add floating announcement bar (dismissible, fades on scroll) - Hero: auto-play chat animation yields to scroll; headline stays static until scroll - Scroll-driven hero phases: chat build → layer separation with Figma-style blueprint annotations (outer/inner borders, guide lines drawing in) → highlight active layer with explanation → reassemble → fade - Bring client logos above the fold with seamless scrolling marquee (compact mode) - Add Greptile and Trigger.dev logos; tune per-logo widths for visual balance - Move hero explanation text to bottom-center at 2xl prominence - Full semantic color pass — replace hardcoded hex with CSS custom properties - Restore footer wordmark, email, and nav links from main branch - Fix video section with real CloudFront MP4 and poster image - Remove tool-calls chat component (unused) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Restructure featured case studies: hero card (Safeway AI) full-width with 3-column row beneath for Cal.ai, Year in Code, Gumloop - Add ghosted company logos as watermarks on all featured cards - Add small logo icons to secondary projects (dRisk, Maige, SyncLinear, Trigger.dev, Greptile) - Introduce --tint CSS variable (#8a9a9a light / #7a9090 dark) as a muted teal accent used on hero category label and hover borders - Update case study data: add category and coPost fields, reorder per spec, update descriptions - Increase card contrast (bg-accent/50 hero, /40 standard) with subtle hover shadows for depth Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…alignment Implement the blog structure spec: split posts into main feed (3 active) and collapsible archive (7 archived), add co-post badges with partner SVG logos (Neon on multi-staging), add editorial bridge note linking multi-staging to Unblocking Agents, add case studies crosslink. Align all blog surfaces (lab listing, post detail, landing preview) with the site's design system — semantic color tokens, border-subtle treatments, and proper light/dark mode hover states. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Create 4 MDX case study files (Safeway AI, Cal.ai, Year in Code, Gumloop) with rich technical content, code blocks, and structured sections - Redesign work listing page with hero card, client logo watermarks, co-published badges (LangChain), and secondary projects section - Polish case study detail pages with category labels, tint accent color, table of contents, and co-post badge pills with logo components - Update featured-work homepage component with matching design language - Add --tint color token to design system (#8a9a9a / #7a9090) - Fix pre-existing TypeScript errors in exploded-view, hero, and session Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Resolves conflicts in case-studies.ts and featured-work.tsx, keeping the MDX-powered versions with richer data model. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…onent expansions Replace v1 dimming/floating-panel approach with v2 architecture: - SVG stroke-dashoffset scaffold overlay with draw-on/erase/ghost animations - Component state machine (active/focused/ghost/hidden) driven by pure function - Expansions grow FROM component edges via connector nodes, not separate panels - Context: sources branch down from question (Safeway, memory, schema, discarded) - Architecture: DAG fans right from reasoning trace (parallel/merge/fallback) - Evaluation: flywheel iterations with improving scores (60→77→91) - 9-phase scroll timeline with collapse transitions between sections - Margin text with project-specific copy (Safeway 250k SKUs, etc.) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Resolves conflict in index.tsx by taking the v2 rewrite (SVG scaffold, state machine, connector-based expansions). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…ctive figures Two companion essays on Rubric's agent architecture thesis: - Contract Engineering: contracts as verifiable specs for AI agents, with interactive contract scenarios slideshow, email app lifecycle mockup showing all 13 contracted events, and animated event verification log - Unblocking Agents: isolation, verification, and persistence for autonomous agents, with terminal mockup showing a blocked agent, resource islands diagram showing isolated agent world vs production, and LLM judge verification loop with screenshot scoring and rework cycle Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…ronger hierarchy Add 5 inline React diagram components (YicFlow, SafewayMemory, SafewaySearchLoop, CalAgentLoop, GumloopTimeline) that visualize each case study's core architecture. Tighten copy across all 4 studies: cut low-signal sections, add code snippets where missing (Cal.ai structured tool, Safeway memory schema), standardize What We Learned format, strengthen outcomes with concrete details, remove redundant footers. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Offset chat panel right on initial load to prevent overlap with "A lab that ships" hero text, then animate to center on scroll. Replace character-by-character typing with input-first flow where the question appears in the input box then submits upward. Tighten expansion layouts (context blocks, DAG nodes, evaluation cards) to fit cleanly on screen at each scroll beat. Tile evaluation iteration cards left-to-right instead of sliding left past the response text. Make video section start paused and grayscale until play is clicked. Normalize card styling across blog, home, and work pages to use consistent bg-accent/40, rounded-xl, and tint hover treatment. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…sign system - Replace Contract Engineering and Unblocking Agents article content with new drafts - Add interactive figures: BeforeAfterFlowFigure (unblocked vs blocked side-by-side), OneWayBridgeFigure (isolation boundary with code diff animation) - Rework PermissionsTrapFigure with sequential line animation and higher contrast - Standardize design tokens across all 10 figure components (7/9/11px font scale, consistent border radius, control icon sizes, transition timing) - Add "New" badges to both posts on lab page and post headers - Move announcement bar to home page only, remove dismissability - Improve article typography: tighter spacing, blockquote sizing, figcaption standardization - Add system prompt and --dangerously-skip-permissions artifacts to Unblocking Agents Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add clickable progress scrubber bars to contract-scenarios, contract-event-log, one-way-bridge, and isolation-architecture figures - Fix blocked timeline: single "Waiting for human" step that stalls for 3 beats instead of repeating "Waiting" three times Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add CopyBlock component with polished hover copy icon - Wrap --dangerously-skip-permissions and system prompt with copy functionality - System prompt copies with redacted keys (****) - Add ... after API keys list - Slow down terminal animation: methodical steps, fast error snap, long pause before yield - Fix system prompt text: "Billing address", "dedicated orgs", "logged in via CLI and browser" - Cut Google/Posthog keys for brevity Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Remove erroneous 'true' from flag (it's a boolean flag, no value) - Update system prompt copy content to use ____ placeholders per spec Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Resolve conflicts by keeping site-rebuild versions (the source branch). Main cherry-picked the unblocking-agents blog post from this branch; conflicts were due to directory renames (src/ui → src/components) and minor editorial differences. Brings in planter.png asset from main. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add non-null assertions for safe array accesses, remove unused imports and variables to satisfy noUncheckedIndexedAccess and strict unused checks. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Restore unblocking-agents files to match main's content (only import path changes from ui→components directory rename remain) - Remove contract-engineering blog post (will live on its own branch) - Update announcement banner and pillars to reference unblocking-agents Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
draft