Deprecate IBC write handlers

[masih]

Now that IBC in/outbound is disabled as part of SIP-3:

• reject all IBC client, connection, and channel messages
• reject IBC transfer messages
• reject IBC client governance proposals
• return stable module deprecation errors
• add coverage for every deprecated write handler

[github-actions]

The latest Buf updates on your PR. Results from workflow Buf / buf (pull_request).

Build	Format	Lint	Breaking	Updated (UTC)
✅ passed	✅ passed	✅ passed	✅ passed	Aug 19, 2026, 9:18 AM

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 58.62%. Comparing base (fab5625) to head (8d9d9b5).
⚠️ Report is 7 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3947      +/-   ##
==========================================
- Coverage   59.60%   58.62%   -0.98%     
==========================================
  Files        2331     2233      -98     
  Lines      200124   188194   -11930     
==========================================
- Hits       119277   110329    -8948     
+ Misses      69413    67401    -2012     
+ Partials    11434    10464     -970     

Flag	Coverage Δ
sei-chain-pr	43.67% <100.00%> (?)
sei-db	70.41% <ø> (-0.22%)	⬇️
sei-db-state-db	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
app/app.go	71.91% <ø> (-0.05%)	⬇️
...ules/apps/transfer/keeper/deprecated_msg_server.go	100.00% <100.00%> (ø)
sei-ibc-go/modules/apps/transfer/module.go	52.77% <100.00%> (ø)
sei-ibc-go/modules/core/02-client/client/cli/tx.go	29.03% <ø> (-1.02%)	⬇️
sei-ibc-go/modules/core/02-client/keeper/client.go	55.73% <ø> (+20.60%)	⬆️
sei-ibc-go/modules/core/02-client/keeper/events.go	58.33% <ø> (+22.33%)	⬆️
sei-ibc-go/modules/core/02-client/keeper/keeper.go	39.69% <ø> (-2.96%)	⬇️
...ei-ibc-go/modules/core/02-client/keeper/metrics.go	50.00% <ø> (ø)
sei-ibc-go/modules/core/keeper/msg_server.go	100.00% <100.00%> (+50.13%)	⬆️
sei-ibc-go/modules/core/keeper/params.go	100.00% <ø> (+50.00%)	⬆️
... and 1 more

... and 102 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[cursor]

PR Summary

High Risk
This permanently blocks live IBC handshakes, relays, and transfers at the message layer—core cross-chain behavior—with a large deletion surface across sei-ibc-go and app wiring.

Overview
Disables new IBC activity by routing all core IBC transaction handlers (client, connection, channel, and packet messages) to return ErrIBCDeprecated instead of executing keeper logic. ICS-20 MsgTransfer is similarly rejected via a dedicated DeprecatedMsgServer, while transfer keeper code remains for versioned EVM precompile historical replay.

Removes governance and CLI paths for IBC client update/upgrade proposals from the main and wasmd apps, and deletes the associated client proposal handlers, keeper methods (CreateClient, UpgradeClient, misbehaviour handling), inbound/outbound gating helpers, and related metrics/events.

Cleans up tests by dropping the wasmd ibctesting harness and IBC relay integration tests, and adding unit tests that assert every deprecated msg server method returns the expected deprecation errors.

^{Reviewed by Cursor Bugbot for commit 8d9d9b5. Bugbot is set up for automated code reviews on this repo. Configure here.}

[seidroid]

The automated review did not complete; see the failing AI Review check for details.

[seidroid]

Clean, well-scoped deprecation of the IBC write paths with good test coverage of the rejecting handlers, but two issues need attention: dropping the ibc gov route can panic the gov EndBlocker on an in-flight legacy IBC client proposal, and rejecting Timeout/TimeoutOnClose/Acknowledgement (which base deliberately left ungated) makes any outstanding packet commitment unresolvable and its escrow unrecoverable.

Findings: 2 blocking | 4 non-blocking | 3 posted inline

Blockers

• None at the file/PR level.
• 2 blocking issue(s) flagged inline on specific lines.

Non-blocking

• Deleting sei-wasmd/x/wasm/ibctesting/*, ibc_reflect_test.go, relay_test.go, and relay_pingpong_test.go removes all integration coverage of the wasm IBC application callbacks in sei-wasmd/x/wasm/ibc.go, which remain in the tree. The new tests only assert the deprecation error, so nothing exercises those callbacks anymore. Worth a note in the PR description that this coverage is intentionally retired along with IBC.
• The tx CLI still advertises commands that can now only produce rejected transactions: sei-ibc-go/modules/core/02-client/client/cli/tx.go keeps create/update/misbehaviour/upgrade, and the transfer module keeps its transfer command. Since the gov-proposal CLI commands were removed in this PR, removing (or marking deprecated in Short) these too would keep the surface consistent.
• sei-ibc-go/modules/core/keeper/params.go still exposes GetParams/SetParams over types.Params{InboundEnabled, OutboundEnabled}, but the core keeper no longer reads either field (only the 03-connection and 04-channel keepers, which read their own copies from the shared subspace). Not a defect, just dead surface on the core keeper now.
• 1 suggestion(s)/nit(s) flagged inline on specific lines.

[seidroid]

[suggestion] The comment says Keeper.Transfer stays executable only for versioned precompiles replaying historical blocks, but the current precompile also calls it directly: precompiles/ibc/ibc.go:169 and :255 invoke p.transferKeeper.Transfer(...), bypassing the msg-service router and therefore DeprecatedMsgServer. So a live EVM IBC.transfer call still attempts a real transfer and fails deeper down with ibc-channel: ibc outbound disabled rather than ErrTransferDeprecated.

Given the PR's goal of a stable deprecation error for transfer writes, consider gating the live precompile too (and adjusting this comment, which currently reads as if only legacy versions reach the keeper).

Blocker 1, acceptable risk at low severity, because it requires a passed IBC proposal to trigger and that category of risk is mitigated by the fact that no proposal can be submitted by the new binary.

Blocker 2, confirmed as a non-issue based on on-chain state.