The state of octez.connect: Who is maintaining this, and why Taquito's migration is blocked

Project Updates
1

Back in February, octez.connect was rushed out in an uncoordinated fork and rebrand of the Beacon SDK, accompanied by an Agora announcement declaring it “Maintained by Nomadic Labs, Trilitech and Functori.”

Anyone running a dApp or wallet at the time remembers how badly that weekend cutover went. Relay nodes were pulled offline without accounting for how the SDK cached server state, locking up roughly a third of ecosystem wallet sessions. While official channels claimed “minimal user disruption” and told end users to manually clear their browser localStorage, we ended up working 15-hour days to diagnose the transport deadlock, stand up @ecadlabs/beacon-*, and ship emergency patches in Taquito just to keep the ecosystem’s wallet connections working.

I maintain Taquito in my free time now. I only ever carried @ecadlabs/beacon-* because upstream was broken, and I want the Beacon maintenance burden off my plate. My plan was to transition @taquito/beacon-wallet to octez.connect once the dust settled and the teams who took ownership of it stabilized their fork.

Instead, eight months later, the stewardship of octez.connect is a mess:

  1. Nobody appears to be at the wheel: When I checked on the octez.connect migration PR in Taquito (ecadlabs/taquito#3435) this week, the Trilitech developer who authored every 5.0.x commit and release replied: “Please get in touch with Vincent poulian. I have moved on.” Meanwhile, Nomadic Labs just announced Octez.js (#7214) on October 21, promoting @tezos-x/octez.js-dapp-wallet on top of octez.connect 5.0.4, seemingly unaware of the state of the package they are wrapping.
  2. Broken core lifecycle and abandoned releases: 5.0.4 (published in July, and still the latest version on npm) cannot even cleanly disconnect and reconnect a wallet in the same browser session. Unaffiliated community contributors have had to diagnose the bugs and submit the fixes themselves (PR #55, Issue #62, PR #63). One fix has been sitting merged on master without an npm release since September 4, and another is sitting unreviewed.
  3. Reckless defaults and neglected security hygiene: A single hardcoded WalletConnect projectId was baked into the SDK as a shared default for all Tezos dApps (PR #50), creating an ecosystem-wide rate-limit and revocation bottleneck while defeating domain verification. Private vulnerability reporting linked in SECURITY.md returns a 404 because advisories were never enabled on the repo (Issue #57), and the SDK still ships with leftover third-party papers.tech telemetry and bug-report endpoints (Issue #59).
  4. No real-world wallet QA: Despite being marketed as the foundation for Tezos X, basic wallet pairing flows on Tezos X Previewnet are broken, featured wallet configuration is broken (Issue #58), and the mobile SDK repositories haven’t been touched since the February rename.

Where Taquito stands

Community contributor @dvkam did the work of cleaning up the Taquito integration in ecadlabs/taquito#3481, and rightly marked the PR [BLOCKED] (“Do not merge it with the current 5.0.4 dependency”) because octez.connect is broken upstream.

I am not going to repeat February by shipping broken wallet infrastructure to Taquito users and spending my weekends cleaning up the fallout, and I am not doing unpaid QA for funded teams.

ecadlabs/taquito#3481 will stay blocked until:

  1. Nomadic Labs, Trilitech, and Functori state who is actually maintaining trilitech/octez.connect.
  2. The open issues and community PRs in trilitech/octez.connect are resolved and published to npm.
  3. The maintainers do their own end-to-end QA across real wallets (Kukai web and iOS, Temple extension and mobile, Umami, and Tezos X previewnet) and get sign-off from the wallet teams confirming that octez.connect v5 is at least as reliable as the @ecadlabs/beacon-* build Taquito ships today.
6 Likes
2

Lack of communication and coordination is not something the agentic age can fix. Please, for the love of God, deal with your shadow and let’s get back on track.

Instead of kicking out Jev, put him in charge. He’s a real one.

If we make this about ego, success will never ever come.

5 Likes
3

Hey, I can see my open bug report from here!

3 Likes
4

there seems to be a trend not just on Tezos but a few other projects as well. this idea that projects don’t need dedicated active maintainers anymore is one that will unfortunately take an active exploit with no one in the radius of blame except management to reverse. the fact that i haven’t been able to reliably land a transaction in over a year without switching wallets constantly doesn’t faze anyone.

as a user i’d find it more respectful if the obviously deprecated projects were made official to reduce the number of rakes we step on

5 Likes
5

Totally agree! But there’s also something to be said about how you move when making projects deprecated.

You can’t have it as your mission statement to provide a good experience for developers and then move like that. Developers on Tezos have not historically had a good experience.

Don’t get me wrong, there are many wonderful things a dapp dev will experience in Tezos, especially if they’re tightly knit with their userbase. But a good/consistent/predictable/sensible/considerate/stable dev experience is not one of the wonderful things.

Putting up some docs on the internet does not a good experience make.

3 Likes
7

Hi Jev,

The developer who was working on octez.connect recently left the team. We respect their decision and wish them all the best. I’m now taking responsibility for coordinating octez.connect going forward, including following up on outstanding issues and contributions.

If you have any specific questions or concerns, I’m happy to discuss them and see how we can move things forward.