What 4 ai coding + dev assistants companies moved this week.

json
{
  "brief_summary": "First-time snapshots of all four AI coding tools (Cursor, Replit, Lovable, Bolt) reveal two distinct wars: Cursor vs. the pro developer with a full agent-OS vision, and Replit/Lovable/Bolt scrapping over the no-code/vibe-coding buyer at near-identical $25/mo price points.",
  "brief_markdown": "# Morthn Intel — AI Coding Tools · Week of June 1, 2026\n\nAll four competitors are new to tracking this week. No diffs — these are baseline reads. Flagging what's strategically loud.\n\n---\n\n## What matters this week\n\n1. Cursor is trying to own the word "agent" at the developer level — and their pricing shows they're betting on it hard. The tier structure runs Hobby → Pro → Pro+ → Ultra for individuals, with a separate Teams/Enterprise ladder. That's unusual segmentation: they're not just upselling seats, they're upselling agent intensity. The FAQ copy explicitly says "we recommend Pro+ for daily agent users, and Ultra for agent power users" — they're training buyers to self-identify by agent appetite, not by team size.\n\n2. Replit, Lovable, and Bolt have independently landed on $25/mo as the magic Pro number. Three competitors, same price point. That's a market signal, not a coincidence. The differentiation fight is now happening inside that $25 — which one has the better token limits, collaboration features, and AI model access. Worth watching who blinks first on price.\n\n3. Bolt is making a serious enterprise bet that the others aren't. Their enterprise page leads with design system integration — "Your company's design system, now in Bolt" — and lists Porsche, Washington Post, and Material UI as examples. That's a different ICP than "next billion creators." Cursor has enterprise controls; Bolt has enterprise positioning.\n\n---\n\n## Cursor\ncursor.com · cursor.com/pricing\n\nHomepage title: "Cursor: The best coding agent." Not an IDE. Not a copilot. An agent. That's a full category-occupation move — they're planting a flag on a word nobody else has claimed cleanly at their level.\n\nThe hero copy: "Built to make you extraordinarily productive, Cursor is the best coding agent." Superlative claim, no hedge. Paired with: "Works autonomously, runs in parallel — Agents use their own computers to build, test, and demo features end to end for you to review." That "their own computers" line is doing heavy lifting — it's positioning cloud agents as a step-change, not a feature.\n\nPricing is the most architecturally interesting of the four. Individual tiers (Pro, Pro+, Ultra) exist alongside a Teams tier at $40/user — meaning they're not collapsing individuals into teams. This is a dual-track go-to-market: win the developer first, then upsell the org. Enterprise adds "AI code tracking API" — which is a new capability worth noting; no competitor mentions anything like it.\n\nWhat to do: If you're competing with Cursor for developer mindshare, "best coding agent" is now the hill they're defending. Positioning yourself as faster, more specialized, or better for a specific language/stack is the only viable counter — don't fight them on the generic agent claim.\n\n---\n\n## Replit\nreplit.com · replit.com/pricing · replit.com/about\n\nHomepage: "Turn ideas into apps in minutes — no coding needed." About page: "We believe software creation should be accessible to everyone, not just programmers." Mission: "Empowering the next billion software creators."\n\nThis is a fundamentally different target customer than Cursor. Replit is not selling to developers — they're selling to people who want to become builders without becoming developers. That's a massive TAM bet.\n\nPricing introduces "Parallel Agents" as a differentiator — Core gets 2, Pro gets 10. That's a clever way to make agent access feel like a resource you grow into, and it creates a natural upgrade driver. The job being hired for here is "ship my idea without hiring an engineer," not "make my engineering faster."\n\nAt $100/mo, Pro is the most expensive non-enterprise individual tier across all four competitors. They're betting that commercial builders will pay 4x the Core price for model access, database rollbacks, and 10 parallel agents.\n\nWhat to do: Replit's "no coding needed" framing is legitimately differentiated from Cursor. If your ICP overlaps with non-technical founders or product managers, Replit is the competitor to watch — not Cursor.\n\n---\n\n## Lovable\nlovable.dev · lovable.dev/pricing\n\nHomepage title: "AI App Builder | Vibe Code Apps & Websites with AI, Fast." They've leaned into "vibe coding" — a term that's memetic but risky to own long-term. It signals speed and approachability, but may not age well as the category matures.\n\nThe pricing structure is unusual: plans are priced per workspace, not per seat — "$25 per month shared across unlimited users." That's a direct shot at Cursor's $40/user Teams plan and Bolt's $30/member Teams plan. If you're a small team, Lovable's math looks dramatically better.\n\nThey gate SSO and "Security center" behind the $50 Business tier — which is smart segmentation. The enterprise tier is "platform fee based on company size, covering all employees," which is a land-and-expand play: get one team in, bill the whole org.\n\nThe reviews page (G2 verified) is the only one of the four competitors actively showing social proof in their tracked pages. Noteworthy quote they're surfacing: "The ability to iterate quickly on ideas without deep technical knowledge is a game changer." Confirms the ICP: technical-adjacent, not pure developer.\n\nWhat to do: Lovable's per-workspace pricing is a competitive wedge worth understanding if you sell to SMB teams. It's a different buying conversation than per-seat.\n\n---\n\n## Bolt\nbolt.new · bolt.new/pricing · bolt.new/enterprise\n\nHomepage: "The #1 professional vibe coding tool." They're claiming the "professional" modifier on vibe coding — trying to own the intersection of approachable and enterprise-ready. The design system angle is their actual differentiator: "Your company's design system, now in Bolt" with named integrations (Porsche Design System, Washington Post Design System).\n\nThis is the most enterprise-credible positioning of the no-code four. The enterprise page leads with a customer quote from a CTO and metrics: "Time to launch reduced from 12 months to 3.5 months." That's not a vibe-coder story — that's a procurement story.\n\nPricing: Free tier is real (300K daily tokens), Pro is $25/mo, Teams is $30/member. The Teams tier adds "Private NPM registries support" and "Design System knowledge with per-package prompts" — features no one else is advertising at this tier. That's a retention play for engineering-adjacent teams who are already in a design system workflow.\n\nThe "98% less errors" claim on the homepage is bold and unattributed. If a competitor can credibly challenge it, that's a vulnerability.\n\nWhat to do: Bolt is the most credible threat to buyers who come from a design/product background and want to ship production-quality code. If that's your ICP overlap, their design system integrations are the feature to benchmark against."
}