feat: consume published verana-types package

[matlux]

Summary

• consume the published @verana-labs/verana-types@0.9.5-pr-268.1 package instead of the local generated codec/Amino implementation
• replace the frontend-owned registry/Amino wiring with createVeranaRegistry() and createVeranaAminoTypes() from the package
• remove the now-unused local app/msg/amino-converter/* layer and switch the remaining codec imports to the package surface
• exclude the local verana-blockchain workspace symlink from TypeScript discovery so local builds do not walk backend test files

Context

This is the frontend companion to verana#268.

The backend work was published as prerelease verana v0.9.5-pr-268.1, which published:

• npm package: @verana-labs/verana-types@0.9.5-pr-268.1
• npm dist-tag: next

That keeps latest on 0.9.4 while giving this branch an honest published dependency to validate against.

Validation

• npx tsc --noEmit
• npm run build

Both passed locally against the published package version above.

[AirKyzzZ]

@matlux Looks good overall

Two things:

1. The dependency is pinned to the prerelease 0.9.5-pr-268.1. What's the plan for bumping to stable once chore(main): release 0.7.0 #268 lands on the backend?
2. Is there a follow-up planned to remove the proto-codecs/ directory entirely afterwards ?

Also be careful there's merge conflicts maybe needs to quick rebase

[genaris]

@matlux Looks good overall

Two things:

1. The dependency is pinned to the prerelease 0.9.5-pr-268.1. What's the plan for bumping to stable once chore(main): release 0.7.0 #268 lands on the backend?
2. Is there a follow-up planned to remove the proto-codecs/ directory entirely afterwards ?

Also be careful there's merge conflicts maybe needs to quick rebase

@AirKyzzZ were you able to test this branch and confirm that verana-frontend works properly (signs transactions, etc.)?

[AirKyzzZ]

@genaris Sure I could test, i'll comment here what i find

[AirKyzzZ]

@genaris Tested the branch locally on Chrome with Keplr wallet, all transactions sign and broadcast successfully via Amino:

• DID creation (MsgAddDID) — OK
• Governance Framework Document (MsgAddGovernanceFrameworkDocument) — OK
• Credential Schema creation (MsgCreateCredentialSchema) — OK

These cover the main type conversion edge cases (simple strings, uint64/Long fields, OptionalUInt32 wrappers). End-to-end migration looks good from the frontend side.

[genaris]

That's good to read! This means that once we merge verana's PR and a v0.9.5 is published, we can slightly update this PR to use that verana-types version and merge it.

…

On Fri, Mar 13, 2026, 12:12 Maxime Mansiet ***@***.***> wrote: *AirKyzzZ* left a comment (verana-labs/verana-frontend#327) <#327 (comment)> @genaris <https://github.com/genaris> Tested the branch locally on Chrome with Keplr wallet, all transactions sign and broadcast successfully via Amino: - *DID creation* (MsgAddDID) — OK - *Governance Framework Document* (MsgAddGovernanceFrameworkDocument) — OK - *Credential Schema creation* (MsgCreateCredentialSchema) — OK These cover the main type conversion edge cases (simple strings, uint64/Long fields, OptionalUInt32 wrappers). End-to-end migration looks good from the frontend side. — Reply to this email directly, view it on GitHub <#327 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABET7TSIDX5RCYBKVA6FNR34QQQM7AVCNFSM6AAAAACWQMZCTOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DANJVHA4TAOJRGI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[antoniobenavidesvallejo]

@AirKyzzZ Of the transactions you mentioned, only the third is representative of the problematic cases. The first one is no longer even used.

It's necessary to test all the credential schema transactions and permission transactions. The permission transactions were the last ones to cause problems, and specific adjustments had to be made to the Amino converters.

[antoniobenavidesvallejo]

If the library includes the amino converter and the protocols indicated in the verana-types issue, it should only be necessary to change the include files and all transactions should work fine.

[matlux]

Yes @antoniobenavidesvallejo , fair point that the perm transactions are the most sensitive Amino cases and that the current frontend checks do not cover all of them yet.

This frontend PR shows that the frontend can consume the published package and that the package swap works end to end. The deeper Amino coverage for credential schema and especially permission transactions belongs on the verana side, because that is the repo that builds and publishes verana-types.

So I agree those are the sensitive cases, but I’d treat that as backend test coverage enhancement rather than something this frontend PR itself needs to solve.