Changelog

Everything that has shipped in Mel, release by release, newest first. During the beta, releases move fast — check back here (or the in-app About settings) to see what changed. For how updates reach you, see Updating Mel.

v0.1.21 — August 2026 (Latest)

• One computer, two accounts, two histories — conversations are stored per account, so signing in as someone else no longer shows you the previous account's conversations. Existing conversations are migrated into place once, automatically.
• Every action answers — a class of buttons used to complete in silence, including Check for updates, which ran a real network check and changed nothing on screen. Actions now confirm, and the ones that fail say so instead of vanishing.
• Menus land where they should — nine dropdowns and pickers were positioned from a height each caller computed by hand; four had it wrong, so lists could open clipped or off-screen.
• The agent can supervise its own work — start a build or dev server in the background, wait for it to be ready, call it over HTTP, open the result in a headless browser to see what actually rendered, and read images. It can also plan first and show you the plan before it changes anything.
• Stuck conversations recover — one malformed message could make every later turn in a conversation fail; Mel now repairs the stored history rather than failing on it.
• Model failover stops sticking — after a provider has a bad minute, the next turn returns to the best available model instead of staying on the stand-in.
• On v0.1.18 or later? The update is one click: avatar menu → Update and relaunch Mel.

v0.1.20 — July 2026

• A real macOS application menu — Mel, File, Edit, View, Find, Window, Help, with the shortcuts you expect (⌘, for Settings, ⌘N, ⌘W). Accelerators are held back while a CLI coding agent has the keyboard, so ⌘P and ⌘D can't be taken from it.
• Window controls behave on macOS — Mel no longer paints its own minimize/maximize/close over the system traffic lights (its copies did nothing in fullscreen). Minimize, Zoom and Toggle Full Screen live in the Window menu.
• Symbols render on macOS — ticks, crosses and close buttons were empty boxes; Mel now ships the font that covers them. The app icon is also sized to Apple's grid, so it matches its neighbours in the Dock.
• The working-directory chip opens a directory picker — in the terminal and in an agent conversation, jump to a parent or subfolder without typing cd.
• Keyboard hints above the prompt are clickable, and two that advertised shortcuts Mel didn't have are gone.
• Starter session templates no longer need a third-party CLI — they open Mel's own agent, and the one that wanted a repository explains itself instead of dead-ending.
• PDF export works on a stock machine — with no browser-based renderer installed, Mel falls back to one it ships with.

v0.1.19 — July 2026

• Blue + pink — Mel's brand is now a true blue, with pink marking everything the agent does (the /agent label, agent sparkles, the running "Melling…" line). Every pairing is contrast-checked.
• Bigger, tunable text — larger by default, plus Settings → Appearance → Font size (13–20pt) that scales the whole app including terminal output.
• One mark everywhere — the app icon is now the same M tile the website uses.
• On v0.1.18? The update is one click: avatar menu → Update and relaunch Mel.

v0.1.18 — July 2026

• One-click updates, for real — earlier builds could never see a new release (their update check pointed at a source that stopped answering); Mel now checks a public release manifest, so the update prompt actually appears when a new version ships. A dot on the avatar signals a ready update, the account menu gained "Check for updates", and a failed check says so honestly instead of pretending you're up to date.
• One-time step — if you're on v0.1.17 or earlier, download v0.1.18 once from the download page; those builds can't update themselves to it. From v0.1.18 on, updating is one click inside Mel. See Updating Mel.
• macOS relaunch fixed — after an update Mel closes and reopens by itself (it used to quit and wait for you in the Dock); updates also survive the move across disk volumes.
• Windows updates go silent — no more installer wizard: Mel closes, updates in the background, and relaunches on its own.

v0.1.17 — July 2026

• macOS is signed and notarized — Mel's macOS build is now signed with an Apple Developer ID and notarized by Apple, so it opens on a normal double-click with no Gatekeeper "Open Anyway" step. Its stable signature also means permission grants (like Full Disk Access) persist across updates. With Windows already code-signed, this is the first Mel release signed on both desktop platforms.

v0.1.13 — July 2026

• Windows code-signing — Windows builds are now code-signed (Concepts Dzign LLC).
• Fixed browser hand-off on Windows — checkout and sign-in links opened without their full address on some upgrades; the browser now receives the complete link on every OS.

v0.1.12 — July 2026

• Accounts on the web — sign up or sign in at openmel.dev, and see your plan, credits, and usage at /account. New in-app Profile and Security pages let you edit your name, change your password, see active sessions and login history, and sign out other devices.
• Sign in with browser — the app can sign in via a short code you confirm on openmel.dev, so you never type a password in the app.
• Real in-app auto-update — Mel downloads, verifies (SHA-256), and applies updates itself, then relaunches.
• Plans in the app now show full feature lists with a monthly/annual toggle, matching the website; plus a contact form at openmel.dev/contact and stronger rate limiting across sign-in and signup.

v0.1.11 — July 2026

• Fixed blank agent transcripts — a long agent conversation could render blank below the fold when scrolled; the full response now stays visible from top to bottom at any window size.

v0.1.10 — July 2026

• Cloud history sync — your agent conversations now sync to your account, so reinstalling Mel or moving to another machine restores your full history instead of starting over.
• Multiple agent chats at once — open several agent conversations side by side in a new Agents sidebar; opening one from history adds it as its own chat instead of replacing the one you're in.
• Agents finish the job — ask the agent to build something and it runs the whole task to completion instead of stopping every few steps to ask permission; a one-key "Keep going" and a hands-off mode make long builds effortless.
• Readable long conversations — very long agent transcripts stay within the window; wide terminal output, file paths, and reasoning are contained instead of being pushed off the side of the screen.
• Running-turn controls — the buttons shown while the agent works now have tooltips and actions: fast-forward the rest of a turn hands-off, or send it to the background and get notified when it's done.
• Windows fixes — agent commands run correctly on Windows; closing a session or tab works on the first click; a not-installed command exits cleanly instead of hanging.

v0.1.9 — July 2026

• Windows installer — a proper per-user setup.exe with desktop and Start-Menu shortcuts and a clean uninstall from Settings → Apps (the portable .zip is still available).
• Readable agent responses everywhere — fixed invisible text and washed-out code cards in the agent transcript on light-themed systems; code blocks are now dark, horizontally scrollable, and copyable.
• Sign-in reliability — fixed a sign-in re-prompt on Windows so your session persists across restarts.

v0.1.8 — July 2026

• Credit plans — every plan includes a monthly bucket of AI credits (1 credit = a cent of real AI cost): Free 150, Pro 1,500, Max 18,000, Business 1,500 per seat. Credits work across every frontier model; bring your own key stays free and unlimited. See Billing & usage.
• Two new plans — Max for power users who live in the agent, and Business with centralized team billing and an org-wide spend limit.
• Add-on credit packs — 400/1k/3k/6.5k packs that roll over month to month and last a year; optional auto-reload with a monthly spend cap you control.
• Self-serve billing — upgrade, buy credits, and manage your subscription from Settings → Billing; a Usage tab shows your per-day credit history.
• Image attachments — attach images to an agent message; image turns route to vision-capable models automatically.

v0.1.7 — June 2026

• Self-serve accounts — the waitlist is gone. Sign up for a free Mel account instantly; the AI agent is free with 150 hosted-AI credits a month or your own key, and a bigger monthly credit allotment on Pro/Max/Business. The website now leads with Download / Get started everywhere. See Account for how sign-up works.
• Enterprise removed — Mel is a personal and small-team tool; the Enterprise/Teams tier and its self-host copy have been removed entirely.
• Pricing simplified — Free and Pro only, two columns, no Teams row. Details in Billing & usage.
• SEO and metadata — every page now has its own OpenGraph/Twitter card, sitemap, and robots.txt; optimised hero video.
• Honesty fixes — removed the fabricated free-credit economy from marketing copy; corrected slash-command names; removed premature open-source claims.

v0.1.6 — June 2026

• Accounts & sign-in — added email/password Mel accounts with an in-app sign-in panel for the agent.
• Account status panel in-app — a clear sign-in/account panel instead of an error when the agent needs authentication. See Common errors for the related auth responses.
• Website redesign — new ink-on-paper design system, real Mel screenshots, provider logos, and a self-playing hero demo video.

v0.1.5

• Agent orchestrator — /orchestrate drives worker Mel agents across panes with cross-agent shared memory. See Orchestration.
• Cross-provider fallback — when a provider is rate-limited or unavailable, the agent automatically reroutes to the next available provider and keeps going. See Models & routing.
• Refreshed catalog — 6 providers, 41 verified models.
• Agent transcript scrolling plus delete-conversation from the landing page.

v0.1.1

• Live command streaming + Stop — watch long commands stream; Stop kills the whole process tree.
• Faster orchestration — the lead agent delegates to workers earlier in a run.

v0.1.0 — Initial public beta release

• Multi-agent orchestration — /orchestrate drives worker CLI agents (Claude Code, Codex) in split panes. See Embedded CLI agents.
• Agent over SSH — melt a remote session into an agent whose tools run on the remote host. See SSH & remote sessions.
• Built-in code review — diffs, inline comments, and LSP-powered navigation. See the Code review panel.
• Command palette + structured blocks — every command becomes a selectable block, and every action is reachable from the command palette.
• Agent mode — 6 providers, many models, no API keys required on your machine.

Found a regression after updating? Tell us — see Getting help.