Comparison

Bystep vs Superset

A desktop “agentic development environment” (Electron) that runs any coding-agent CLI — Claude, Codex, Cursor, OpenCode, Gemini and more — in a terminal per git worktree. It uses the subscriptions and API keys you already have and, in its own words, “never proxies model calls”. macOS first; the Linux build is experimental; no Windows yet.

What Superset is

A desktop “agentic development environment” (Electron) that runs any coding-agent CLI — Claude, Codex, Cursor, OpenCode, Gemini and more — in a terminal per git worktree. It uses the subscriptions and API keys you already have and, in its own words, “never proxies model calls”. macOS first; the Linux build is experimental; no Windows yet.

Pricing
Free · Pro $20 ($15 on yearly billing) · Enterprise — as listed on superset.sh/pricing when we checked on 10 Sep 2026. Check their page for current numbers.

Competitor facts checked 10 Sep 2026 · Sources: superset.sh/pricing, github.com/superset-sh/superset

Feature matrix

Where the two actually differ

Every Bystep item here is live in the product today — not a roadmap promise. Competitor claims we couldn't verify first-hand are marked “as far as we know.”

BystepSuperset
PRD generationYesNoStarts from a worktree and a terminal, not from an idea-to-PRD flow
Acceptance criteria per taskYesids like AC-checkout-1, derived from the PRD, editable by handNoNo structured acceptance-criteria field per task, as far as we know
Verification gateYes`task complete` returns 409 without passing evidence or a recorded overrideNoNothing requires evidence before a task counts as done
Evidence panelYestest output, diff stat, files touched, optional screenshot — stored, not just claimedNo
Works with any coding agentYesClaude Code, Cursor, Codex, Copilot, OpenCode, Gemini CLI via CLI + MCPYesIts defining trait: Claude, Codex, Cursor, OpenCode, Gemini, Kimi, Grok, Hermes and more
Import / export formatsYesSpec Kit, OpenSpec, Backlog.md, Task Master, Kiro, Claude Code Tasks, GitHub Issues, LinearNoLinear and Slack integrations, not a spec/task format hub
Client mode / proposalYesclient-viewer share page + proposal for the Studio tierNo
Cost per featureYesper feature, not just per sessionNoNot shown per feature, as far as we know
Indonesian language & paymentsYesen/id UI, IDR pricing, QRIS/VANoEnglish product, USD pricing
Open-source CLIYes`@bystep/cli` + `@bystep/mcp`, MITPartialSource-available under the Elastic License 2.0 — readable, but not OSI open source

Honest answer

When to choose Superset instead

  • You are a developer who wants one desktop app to drive many different agent CLIs side by side, each in its own terminal and worktree.
  • You need remote access, Linear/Slack integrations and a vendor with SOC 2, and you want to be able to read the source.
  • Your tasks are engineering work you verify yourself — you don’t need a PRD, a verification gate, or a page to show a non-technical stakeholder.

Keep your agents, add the proof

Bystep runs the agents you already use from one plan, then requires evidence before a task counts as done. The result is readable by people who don’t read diffs.