Spaces:
Running
Running
Commit History
15 days ago
-
feat(ui): teal + gold for the agent UI and the scored result
Claude +1: Claude Sonnet 5.5Unverified signatureAgent - Strong job cards get a gold seal: gold edge and glow, gold-ringed score with a teal-to-gold percentage. Verdict colour stays green. - Active chat in the session rail has a gold edge; "Chats" label gets the gold diamond; New chat hovers gold. - User chat bubbles get a teal outline. Scored result - Strong match percentage runs teal to gold (with padding so the % is not clipped); the next-step quote gets a teal-to-gold spine. - The best-scoring dimension is marked with a gold diamond. Checked job cards and result hero in a headless browser with sample data. Logo, brand assets and both brand hex values are unchanged. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Q54vp7U5MJ3JrbTQk45zmta520b27
-
feat: add premium loading motion across scorer and agent
Cursor Agent +1: Zimuzo EjinkeonyeUnverified signatureGive scoring, agent thinking, and internal loading states a shared night-signal treatment—orbiting desk mark, scan rails, and score reveals—without changing theme, logo, or behavior. Co-authored-by: Zimuzo Ejinkeonye <zimejin07@users.noreply.github.com>97f9abe -
feat(search): score every kept listing and show which read it was
Cursor Agent +1: Zimuzo EjinkeonyeUnverified signatureThe model still reads the top shortlist. Listings past that are checked against the profile in the same pass instead of being dropped, and the card says Full score or Quick check. A search now costs 300 credits. Co-authored-by: Zimuzo Ejinkeonye <zimejin07@users.noreply.github.com>635aaf9
-
feat(ui): bring the agent stage to life
Cursor Agent +1: Zimuzo EjinkeonyeUnverified signatureAdd a living signal mark on the empty agent view, search-light and arrival motion for messages, matches, and scores, and keep the existing logo and palette. Co-authored-by: Zimuzo Ejinkeonye <zimejin07@users.noreply.github.com>92b63b1
-
feat(ui): night signal desk retheme across marketing and product
Cursor Agent +1: Zimuzo EjinkeonyeUnverified signatureReplace purple-glow SaaS look with teal-ink + lamp-gold tokens, Fraunces display type, brand-first full-bleed landing hero, and aligned chrome, agent cards, dashboard, and shortlist surfaces. Co-authored-by: Zimuzo Ejinkeonye <zimejin07@users.noreply.github.com>8e51d1a
-
fix(discovery+ui): reject aggregator hubs and polish match cards
Drop Matcha/Indeed/Fossjobs-style index pages before scoring, keep ATS-first angles when the planner LLM is down, and tighten card hierarchy so chat and dashboard feel consistent. Co-authored-by: Cursor <cursoragent@cursor.com>a360cf0 -
Improve job discovery strategy and result cards
Co-authored-by: Cursor <cursoragent@cursor.com>c708e76
-
Tell the truth about match percent, verdict, and what we search.
Co-authored-by: Cursor <cursoragent@cursor.com>f944684 -
Lead scored jobs with a match percent, not only a verdict.
Co-authored-by: Cursor <cursoragent@cursor.com>57e740d -
Make Agent a single stage: drop classic mode and keep chips off the composer.
Co-authored-by: Cursor <cursoragent@cursor.com>c54e3de
-
Quiet the Agent composer and loading states without changing the search chips.
Co-authored-by: Cursor <cursoragent@cursor.com>bb6603a -
Let users rename chats and keep drafting without losing their place in the thread.
Co-authored-by: Cursor <cursoragent@cursor.com>d114205 -
Idle the agent until the user sends a message, like ChatGPT and Claude.
Stop auto-starting an init run on page load so New chat can switch immediately without hitting Stop, and abort the previous stream on unmount. Co-authored-by: Cursor <cursoragent@cursor.com>f6127c3
-
Unify signed-in chrome into a thin app bar and give Agent the remaining viewport.
Co-authored-by: Cursor <cursoragent@cursor.com>cf60152 -
Switch the UI to a warm ink-studio dark theme.
The washi-light palette was high-contrast and fought dark OS preference and the overnight-agent product. Co-authored-by: Cursor <cursoragent@cursor.com>26af279
-
Saturation pass, nav label unification, chat chip/stop polish, discovery filter diagnostics
Color confidence (visual only): - primary-light #ebe9f8 -> #ded9f5; card border 0.07 -> 0.10; border-strong 0.16 - add terracotta secondary accent (#c2542f) tokens; use on pricing label/highlight - give signal/verdict/accessibility pills real borders instead of borderless tints - retune leftover blue/amber rgba literals to the indigo/terracotta palette Nav labels unified to Dashboard / Agent / Job Scorer / Job preferences: - 'Job Agent' -> 'Agent', 'My Findings' -> 'Dashboard', drop 'Back to' prefix - consistent order across DashboardNav, agent, batch/crawl/scrape, home, shortlist Agent chat: - Stop button is now ghost/secondary with danger-tinted hover - suggestion chips are single-line, width-capped, ellipsised, with title tooltips (display text only; full label is still the message payload) Discovery diagnostics: - log per-angle Hirebase queries plus drop counts and a 10-row sample with the accessibility classification behind each exclusion - refresh stale 'Global remote onl…eba4dee
-
Bold UI refresh with shared tokens and CSS-only surface updates.
Refresh marketing, agent, dashboard, and ScoreBar styling while keeping routes and behavior unchanged. Co-authored-by: Cursor <cursoragent@cursor.com>e0402df
-
Add durable search runs with resumable agent execution.
Introduce Redis-backed SearchRun checkpoints, QStash workers, session transcript restore, and reconnectable chat UX so discovery can resume after failures without redoing completed work. Co-authored-by: Cursor <cursoragent@cursor.com>bcdd291
-
Upgrade discovery yield, digest reliability, and runtime UI.
Improve search pivots, heuristic scoring fallback, unified global-remote rubric, segmented seen-url TTLs, feedback loops from save/dismiss/apply, and live agent status on dashboard and chat. Co-authored-by: Cursor <cursoragent@cursor.com>b7bf6e2
-
Agent chat: OpenRouter tool-model allowlist, discover UX, plan card polish
Co-authored-by: Cursor <cursoragent@cursor.com>ae7c9a0
-
Add agent observability (Redis run logs, lab debugger) and copy session export
Persist agent runs to Redis with optional verbose LLM traces when LAB_LOG_INFERENCE is enabled. Instrument runAgentConversation with completion steps and tool timeline; expose X-Agent-Run-Id on chat responses. Add lab GET /api/lab/agent-runs and AgentRunsDebugger UI. Lab admins get Copy debug trace in AgentChat; all users keep markdown Copy session export. Co-authored-by: Cursor <cursoragent@cursor.com>888d34f
-
Harden agent: discover timeout hints, batch marker UI, errors, blocked domains
Co-authored-by: Cursor <cursoragent@cursor.com>88fc77f -
Improve batch search UX: batch_started card, guards, check_batch fix, discover intent
Co-authored-by: Cursor <cursoragent@cursor.com>372300b
-
Cap discover_jobs LLM scoring to 10, timeout guard, Stop stream, vector warn
- Pre-filter with Upstash vector (top 10) or title-keyword heuristic; score at most 10 listings per call; update partial results after each batch for timeouts. - Block repeat discover_jobs in the same request after timedOut; slim LLM tool payload for discover_jobs; align system prompt and mission reminder. - Agent chat: AbortController on fetch and NDJSON read; muted Stop button while busy; init uses same abort pattern. - Log once when UPSTASH_VECTOR_REST_URL is unset. Co-authored-by: Cursor <cursoragent@cursor.com>a5a8dee
-
feat: implement core agent chat framework with tool execution, schema definitions, and initial UI component.
fb09452
-
Improve profile panel gating, subscription messaging, and marketing landing.
- JobApp: scoring context shows CV-backed content only when experience summary and strong skills exist; empty state with Upload CV opens onboarding; Scoring as line only with summary plus roleType from preferences; remove demo placeholder identity. - JobApp: non-subscriber scoring gate uses value-forward copy, Stripe checkout CTA, and calm card styling. - AgentChat: replace red subscription gate with neutral card, new headline and body copy, priced CTA and cancel note. - Marketing: animated hero agent preview, gradient hero background, CTA pulse, timeline steps with SVG icons, sample score bars, placeholder testimonial cards (TODO), premium pricing card with regional note, larger footer CTA, increased section spacing. Made-with: Cursorcc26136
-
Move digest status into agent chat; remove launcher card
Made-with: Cursordd8fff1
-
feat(agent-chat): subscribe CTA when subscription inactive
Show a gated panel with Stripe subscription checkout for signed-in users without subscriber inference access; skip init request and disable composer until entitled. Styles for subscribe gate and disabled saved chip. Made-with: Cursor08942b1 -
feat(agent): proactive planning agent with init snapshot and tool budget
- API: init flag, session snapshot from shortlist + credit usage - Runner: plan-first for search-like intent, MAX 10 tool calls, progress/plan NDJSON - Groq/OpenRouter: optional tool_choice none for init and summaries - UI: mount init stream, plan checklist, progress rows; remove hardcoded welcome Made-with: Cursor6e665d0 -
feat(agent-chat): add credit top-up action pills
Emit a structured top-up action on insufficient credits and render it as a pill CTA in chat messages. Made-with: Cursor47a4568
-
feat(job-launcher): agent chat UI + API; refine Job Scorer home layout
- Add /api/agent/chat NDJSON stream with Groq tools (OpenRouter fallback) and same-origin calls to existing launcher/score/shortlist routes. - Add AgentChat, AgentJobCard, context strip, and lab admin Agent/Classic toggle. - Add lib/agent-chat helpers (search_jobs via Serper + generateSearchStrings). - Extend ShortlistSaveButton emphasize prop; shortlist hook savedCount. - Rate-limit /api/agent/chat in middleware. - Job Scorer: agentic profile panel, score shell, batch below results; main landmark. - globals: job-app and agent UI tokens, visually-hidden, result enter animation. Made-with: Cursorc78b3b6