Sulu
Sulu ships 2.6 and 3.0 in lockstep, and the 3.0 line still reads like a maintenance branch.
A side-by-side editorial comparison of n8n and Kit (formerly ConvertKit) — release velocity, themes, recent moves, and the top alternatives to consider.
n8n's daily patch train hides the real work: hardening its MCP server into a proper auth resource.
n8n ships patches almost daily across three live branches (2.31.x, 2.32.x, 2.33.x), so the changelog reads mostly as noise. Underneath it, nearly every substantive commit lands in one of two places: the AI agent builder and the MCP server. The 2.33.0 minor is the clearest example, a very large bundle with no features section at all, only fixes.
Kit rebuilt its landing page editor, then handed the controls to an AI agent.
Kit's recent releases converge on one thing: giving creators a full publishing surface inside Kit rather than sending them to a third-party page builder. The landing page editor was rebuilt in June with 20+ templates and full control over layout, fonts and sections; custom favicons and the Creator profile to Newsletter site rename finished the job of making those pages feel owned rather than hosted. The newest release lets the Kit MCP build and edit those pages directly. Alongside runs an audience-intelligence thread - Subscriber Signals in early access - and network features letting creators request connections with each other.
n8n ships patches almost daily across three live branches (2.31.x, 2.32.x, 2.33.x), so the changelog reads mostly as noise. Underneath it, nearly every substantive commit lands in one of two places: the AI agent builder and the MCP server. The 2.33.0 minor is the clearest example, a very large bundle with no features section at all, only fixes.
The MCP work has moved past merely exposing a server toward protocol compliance: OAuth protected-resource metadata served at the bare well-known path, the RFC 9207 iss parameter on authorization error redirects, PKCE negotiation and token expiry handling. Agent-side changes point the same direction, with secrets scrubbed before they reach agent memory, MCP connection failures made non-blocking for agents, and workflow deletion refused while an unpublish is still running. This is the unglamorous half of making agents safe to run inside someone else's instance.
With the auth and audit plumbing largely landed, the next visible move is agent capability that depends on it, such as delegated MCP access, rather than another round of protocol fixes.
Kit's recent releases converge on one thing: giving creators a full publishing surface inside Kit rather than sending them to a third-party page builder. The landing page editor was rebuilt in June with 20+ templates and full control over layout, fonts and sections; custom favicons and the Creator profile to Newsletter site rename finished the job of making those pages feel owned rather than hosted. The newest release lets the Kit MCP build and edit those pages directly. Alongside runs an audience-intelligence thread - Subscriber Signals in early access - and network features letting creators request connections with each other.
The Kit MCP is the axis. It arrived in May as a read-and-analyze tool - ask questions about your list, create tags or broadcasts from Claude, ChatGPT or Cursor - and has now crossed into building the actual published artifact. That sequencing is deliberate: rebuild the editor as a structured, template-driven system first, then expose it to an agent, because an agent can only assemble a page that is composed of describable parts. The rename and favicon work point the same way, toward Kit as the place a creator's web presence lives rather than just where email is sent.
Expect the MCP surface to extend from landing pages into sequences and the newsletter site itself, and Subscriber Signals to move out of early access as the data layer the agent reasons over.
Other Mkt Auto products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either n8n or Kit (formerly ConvertKit).
Sulu ships 2.6 and 3.0 in lockstep, and the 3.0 line still reads like a maintenance branch.
Nearly every release in this window carries security hardening — Pimcore is working through an injection sweep.
The v4 canary line is being renamed Build Awesome while 3.x is kept on pure maintenance.
Steady feature-then-patch cadence, with security reports rising because AI tools are finding them
Measurement gets granular while the API quietly opens ClickFunnels up to outside agents
OneSignal is arguing a new category into existence before showing the product behind it.
See all n8n alternatives → · See all Kit (formerly ConvertKit) alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
Both compete on the same themes — mcp — within Mkt Auto. Kit (formerly ConvertKit) is currently shipping more aggressively (velocity 6.3 vs 5.0), with 1 editorial sparks in the last 30 days against 0. See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. Kit (formerly ConvertKit) is currently shipping more aggressively (velocity 6.3 vs 5.0), with 1 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other Mkt Auto products to evaluate alongside.
Top n8n alternatives in Mkt Auto are ranked by recent ship velocity. Browse the "n8n alternatives" section above for the current picks, or visit /alternatives/n8n for the full list with editorial commentary on each.
Top Kit (formerly ConvertKit) alternatives in Mkt Auto are ranked by recent ship velocity. Browse the "Kit (formerly ConvertKit) alternatives" section above for the current picks, or visit /alternatives/convertkit for the full list with editorial commentary on each.