Fix Cargo source replacement being ignored (#338, #455) by mikolalysenko · Pull Request #1420 · SocketDev/socket-patch

Mikola Lysenko (mikolalysenko) · 2026-10-12T06:24:48Z

LLM Description written by Claude Code:claude-opus-5-5

Fixes #338
Fixes #455

Summary

socket-patch never read Cargo's source replacement
([source.crates-io] replace-with = "…" → [source.<name>] directory = "…"),
so it guessed where a cargo vendor project builds from, and guessed wrong.

Agent mode (Agent-mode cargo apply patches the wrong copy when cargo vendor uses a custom directory (or apply runs from a workspace member), yet reports success and VEX attests not_affected #338, Hosted cargo scan in a cargo vendor project reports success but breaks every fresh cargo build --frozen --offline, and VEX then omits the patch #455 symptom 2). The crawler hard-coded
<cwd>/vendor. With a custom vendor directory, with a workspace-root
vendor directory seen from a member, or with an unwired or non-cargo
vendor/, apply patched a copy cargo doesn't build from and reported
success, and vex judged the wrong copy. The crawler now follows the
replace-with chain of every config cargo reads (each ancestor's
.cargo/config*, then $CARGO_HOME's) and crawls the directory source it
ends on, resolved the way cargo resolves config paths (against the parent
of .cargo/). Without a directory replacement it crawls the registry
cache. A configured directory that is missing yields nothing, never a copy
cargo ignores.
Rollback/prune (Cargo rollback after an agent→vendored takeover leaves the shared registry cache patched, reports success, and deletes the revert blobs #336 leg). find_cargo_copies decided that a vendor
directory hides the registry cache by checking <cwd>/vendor. It now asks
the crawler for the configured directory source.
Hosted mode (Hosted cargo scan in a cargo vendor project reports success but breaks every fresh cargo build --frozen --offline, and VEX then omits the patch #455 symptom 1). When a config cargo reads for a build
in the project (its own .cargo/config*, a parent directory's such as a
monorepo root, or $CARGO_HOME's) replaces crates.io with an on-disk
source (a directory source or a cargo local-registry), every cargo
dependency is skipped with nothing written and a new additive
redirect_cargo_directory_source warning that points at agent mode. The
CLI resolves this read-only and passes it into the pure rewriter, as
yarn's outer mirror is passed. A registry-mirror replacement is still
redirected.

Behaviour change to note: a vendor/ that no replace-with wires is no
longer crawled, patched, rolled back or pruned. Cargo never built from it, so a
patch an earlier release wrote there had no effect. Rollback and prune now look
at the copy cargo does build. A stray edit an earlier release left in such an
unused vendor/ stays where it is; restore it from version control if needed.

The new pure module formats/cargo/source_replacement.rs holds the resolver
(crates_io_source) and the config file walk (config_layer_paths).
CLI_CONTRACT.md and docs/ecosystems.md are updated.

Per-issue checklist

Issue	Regression tests (red before the fix → green)
#338 shape 1 (custom dir)	`crawler_cargo_e2e::custom_vendor_dir_named_by_source_replacement_is_crawled`; real cargo: `e2e_safety_cargo_build::apply_patches_a_custom_named_vendor_dir`
#338 shape 2 (member)	`crawler_cargo_e2e::workspace_root_source_replacement_applies_to_a_member`; real cargo: `e2e_safety_cargo_build::apply_from_a_member_patches_the_workspace_vendor_dir`
#338 shape 3 + non-cargo `vendor/`	`crawler_cargo_e2e::unwired_vendor_dir_falls_back_to_the_registry_cache`
#338 fresh checkout / missing dir	`crawler_cargo_e2e::missing_replacement_directory_crawls_nothing`
#338 `$CARGO_HOME` config	`crawler_cargo_e2e::cargo_home_source_replacement_is_honoured`
#455 hosted redirect	`patch::redirect::tests::cargo_directory_source_replacement_is_refused` (both config spellings, CRLF, a multi-hop chain; mirror control still redirects)
#455 vex reads the wrong copy	covered by the crawler change: vex discovers through the same crawler (`vex_consumed.rs`); the crawler tests above
resolver	`formats::cargo::source_replacement::tests` (8 unit tests)

Red before the fix: all 5 new crawler_cargo_e2e tests failed on origin/main
code. The hosted test failed with the Socket registry pin written into
Cargo.toml, Cargo.lock and .cargo/config.toml.

Fixtures that staged a bare vendor/ and relied on the old guess now write the
.cargo/config.toml that cargo vendor asks for (6 crawler unit/oracle tests,
crawlers_empty_paths_e2e, e2e_cargo ×2, e2e_vendor_cargo_build's #336
"behind a cargo vendor dir" rollback). None of their assertions were loosened.

Local evidence (on `30aa879`; `9083269` adds 3 fixture wirings, CI green)

cargo fmt --all -- --check: clean
cargo clippy --workspace --all-features -- -D warnings: clean
cargo test -p socket-patch-core --all-features: all pass except 4
read-only-permission tests that can't fail when run as root in this sandbox
(copy_tree::relax_loop_must_not_traverse_symlinked_root,
vlt_heal::an_unremovable_hidden_lock_keeps_every_store_entry,
pypi_poetry::wire_write_failure_…, pypi_requirements::wire_failure_…).
These are unrelated to this diff.
socket-patch-cli: --lib, e2e_cargo, in_process_cargo_apply,
mode_migration_cargo, e2e_redirect_cargo_build, e2e_redirect_cargo_shapes,
e2e_vendor_cargo_build (incl. both Cargo rollback after an agent→vendored takeover leaves the shared registry cache patched, reports success, and deletes the revert blobs #336 rollback tests with real cargo),
vex_terminal_output, in_process_rollback_hosted, in_process_vendor,
covgap_commands_scan_mod: pass
cargo test -p socket-patch-cli --all-features --test e2e_safety_cargo_build -- --ignored:
10/10 pass with real cargo 1.93.1, including the two new Agent-mode cargo apply patches the wrong copy when cargo vendor uses a custom directory (or apply runs from a workspace member), yet reports success and VEX attests not_affected #338 shapes
The full cargo test --workspace couldn't finish locally: the sandbox ran
out of disk building every CLI test binary. CI runs it.
Wrappers (npm/, pypi/, gem/) only dispatch to the binary, so no
parallel change is needed.

/code-review high: findings not fixed here

Vendored mode's is_vendored (vendor/cargo.rs) still probes
<root>/vendor for its already_vendored_in_tree refusal. That is a
fail-closed refusal with its own contract tests, and neither issue
reports it. Follow-up.
Upgrade from a release that patched an unwired vendor/. Cargo never
built from that copy, so the patch had no effect. rollback now looks
where cargo builds. The stray edit in the unused vendor/ copy stays.
Registry mirror / local-registry ordering. With a mirror or local-registry replacement, when both the
crates.io and the mirror unpack dirs hold the crate, the crawler takes the
first in listing order. That predates this PR: main crawls every
registry/src/* dir the same way. Choosing the mirror's dir means
reproducing cargo's SourceId hash. Follow-up.
Per-key (not per-table) merge of conflicting [source.x] definitions:
cargo itself rejects those ("more than one source location"). A .. in
--cwd is folded before the walk (ae078c1, b583db6 keeps the given cwd as the reported prefix). A symlinked cwd is walked
logically rather than physically: absolute() is used rather than
canonicalize() so reported paths keep matching the cwd the user gave
(macOS /var → /private/var).
Env-var and --config source replacement (CARGO_SOURCE_*) is not
read in either mode.
A third config-chain walker / no memo. vendor::cargo_config::read_config_chain
is [patch]-specific (it drops files with no patch entries). Unifying the
walkers is a refactor for its own PR. The extra reads are at most a handful
of small files per crawl.

🤖 Generated with Claude Code

https://claude.ai/code/session_01TUHnkhA7W5gHh21jSMkY7P

Generated by Claude Code

Assisted-by: Claude Code:claude-opus-5-5

Agent-mode apply, scan and vex looked for a `cargo vendor` tree only at `<cwd>/vendor`. A custom vendor directory, a workspace root's vendor directory seen from a member, or a `vendor/` that no config wires all made apply patch a copy cargo ignores while reporting success, and made vex judge the wrong copy. The crawler now follows the `[source.crates-io] replace-with` chain of every cargo config cargo reads (each ancestor's `.cargo/config*`, then `$CARGO_HOME`'s) and crawls the directory source it ends on, resolved the way cargo resolves config paths. Without a directory replacement it crawls the registry cache, and a configured directory that is missing yields nothing rather than a copy cargo doesn't read. Refs #338, #455 Assisted-by: Claude Code:claude-opus-5-5

A project whose .cargo/config replaces crates.io with a directory source (`cargo vendor`) builds every crate from that directory and never reads a registry. `scan --mode hosted` still pinned the crate to a per-patch Socket registry and reported success, so every fresh `cargo build --frozen --offline` failed with "no matching package". The hosted cargo rewriter now follows the project config's `[source.crates-io] replace-with` chain and, when it ends on a directory source, skips every cargo dependency with nothing written and a `redirect_cargo_directory_source` warning that points at agent mode. A registry mirror replacement is still redirected. Refs #455 Assisted-by: Claude Code:claude-opus-5-5

Adds real-cargo e2e coverage for a custom-named vendor directory and for apply run from a workspace member, and documents how agent mode finds the directory source and the new hosted refusal code. Refs #338, #455 Assisted-by: Claude Code:claude-opus-5-5

Three fixtures staged a bare `vendor/` with no source replacement and relied on the crawler treating it as cargo's source. They now write the `.cargo/config.toml` that `cargo vendor` asks for, and the resolver strips a BOM through the shared text helper. Refs #338 Assisted-by: Claude Code:claude-opus-5-5

Rollback and prune look for an agent-mode patch left in the shared registry cache when the project crawl searches only a `cargo vendor` directory. That check still looked for `<cwd>/vendor`, so a custom or workspace-root vendor directory skipped it, and an unwired `vendor/` triggered it needlessly. It now asks the crawler for the directory source the project's cargo configs name. The hosted refusal's message now says why a Socket registry pin breaks an offline vendored build, instead of claiming cargo never reads it. Refs #338, #455 Assisted-by: Claude Code:claude-opus-5-5

Mikola Lysenko (mikolalysenko) · 2026-10-12T07:33:57Z

BugBot review

Generated by Claude Code

cursor

Stale Bugbot comment from a previous run.

The dispatch, all-ecosystem rollback and scan-scope tests staged a bare `vendor/` crate tree and relied on the crawler treating it as cargo's source. They now write the source replacement `cargo vendor` asks for, so they keep exercising the vendored-crate path. Refs #338 Assisted-by: Claude Code:claude-opus-5-5

Mikola Lysenko (mikolalysenko) · 2026-10-12T07:53:21Z

BugBot review

Generated by Claude Code

cursor

Stale Bugbot comment from a previous run.

Mikola Lysenko (mikolalysenko) · 2026-10-12T09:11:59Z

Ready for review (burn-down agent)

Head: 9083269
CI: 35/35 non-skipped checks green (19 skipped by path filters); mergeable, no conflicts, no CHANGELOG.md change.
Bugbot: reviewed 9083269, no issues found; no open review threads.
Reviewer focus: .cargo/config(.toml) source replacement is now read, so the crawler stops assuming <cwd>/vendor and the hosted redirect respects directory sources (Agent-mode cargo apply patches the wrong copy when cargo vendor uses a custom directory (or apply runs from a workspace member), yet reports success and VEX attests not_affected #338, Hosted cargo scan in a cargo vendor project reports success but breaks every fresh cargo build --frozen --offline, and VEX then omits the patch #455).

Generated by Claude Code

Mikola Lysenko (mikolalysenko) · 2026-10-12T10:03:18Z

Final review brief

What it does: socket-patch now reads Cargo's source replacement ([source.crates-io] replace-with → [source.X] directory), following the chain through every ancestor's .cargo/config* and then $CARGO_HOME. Agent mode patches the directory cargo actually builds from, not a hard-coded <cwd>/vendor, and rollback and prune use the same lookup. Hosted mode skips cargo pins with a new redirect_cargo_directory_source warning when crates.io is replaced by a directory source.

Risk: low to medium. The resolver is small, pure and well tested: cycles and malformed TOML fall back to the registry, with no panic or loop. The risk is the intended behaviour change: a vendor/ that no replace-with wires up is no longer crawled or patched, and a configured directory that is missing crawls nothing.

Look here:

source_replacement.rs#L51-L125:`` chain resolution, layer precedence, and config winning over `config.toml` (as Cargo does)
cargo_crawler.rs#L282-L331:``` the new root selection, and the std::path::absolute choice
redirect/mod.rs#L1736-L1763:```` the hosted early return
ecosystem_dispatch.rs#L104-L115: the rollback/prune gate for a shadowed registry copy
crawler_cargo_e2e.rs#L858-L1025:````` the new Agent-mode cargo apply patches the wrong copy when cargo vendor uses a custom directory (or apply runs from a workspace member), yet reports success and VEX attests not_affected #338 scenarios

Verified:

Read the full diff and checked it against Cargo's config rules: precedence, config-relative directory paths, cycles, CRLF/BOM, and CARGO_HOME unset.
Ran locally, all passing: crawler_cargo_e2e (39), crawlers_empty_paths_e2e (13), the source_replacement, crawler and hosted-refusal unit tests (44), and the two new real-cargo e2e_safety_cargo_build tests (2/2; cargo check --frozen --offline succeeds on the patched vendor copy).
CI ci-ok and clippy are green on 9083269, and Bugbot reports no open findings.
No unresolved threads, and CHANGELOG.md is untouched.

Changes I made: none.

Open questions (none block merge; they could be follow-ups):

A cwd containing .. (e.g. --cwd ../proj) walks the configs of directories that aren't ancestors, because std::path::absolute keeps .. on Unix. A probe with T/b/../proj picked up T/b/.cargo/config.toml. A symlinked cwd differs from Cargo's physical-path walk in the same way. A lexical .. cleanup before the walk would fix it.
Hosted mode refuses only directory replacements. A local-registry replacement is just as offline and is still pinned. Should it also be refused, or at least warned about?
Hosted mode reads only the repo-root .cargo/config*, so a replacement set in $CARGO_HOME or an ancestor above the repo isn't seen. The env-var and --config forms (CARGO_SOURCE_CRATES_IO_REPLACE_WITH, …) are ignored in both modes. These may be worth a line in CLI_CONTRACT.md.
The PR description could call out the behaviour change for an unwired vendor/, which earlier releases patched. Without a replace-with, rollback and prune no longer see those copies.
Test nit: the crawler unit tests and some CLI e2e tests now read the developer's real $CARGO_HOME/config.toml. Pointing CARGO_HOME at a tempdir, as CargoHomeGuard already does, would make them hermetic.

Auto-merge is armed, so approving sends this straight to the merge queue.

Generated by Claude Code

A --cwd such as `b/../app` walked `b`'s ancestors and could pick up a sibling directory's `.cargo/config.toml`, so the crawl followed a source replacement cargo never sees from `app`. The path is folded lexically before the walk. Refs #338 Assisted-by: Claude Code:claude-opus-5-5

Mikola Lysenko (mikolalysenko) · 2026-10-12T10:11:45Z

[agent] Thanks for the brief. On the open questions:

.. in --cwd: fixed in ae078c1. The cwd is folded lexically before the config walk, and a new regression test, parent_dir_in_cwd_does_not_read_a_sibling_config, fails without the fold. A symlinked cwd is still walked logically. canonicalize() would make reported paths diverge from the user's cwd (macOS /var → /private/var), so I left that as is.
local-registry in hosted mode: not refused here. Cargo unpacks local-registry crates into registry/src, so agent mode is correct. Whether a hosted pin breaks there depends on whether the Socket registry is reachable at build time, which is a separate decision. I'd rather track it as a follow-up than widen this PR.
Hosted mode reads only the project config; env/--config forms are ignored: both are now listed in the PR description under the review notes. Wiring outer configs into the hosted engine, as yarn's OuterYarnMirror does, is a follow-up.
Behaviour change for an unwired vendor/: now called out in the PR description.
Test hermeticity: the new tests use CargoHomeGuard. The older crawler unit tests and CLI fixtures could read a developer's ~/.cargo/config.toml, but only a crates.io replacement there would change their result. Making them hermetic is a test-only follow-up.

Generated by Claude Code

Mikola Lysenko (mikolalysenko) · 2026-10-12T10:11:46Z

BugBot review

Generated by Claude Code

cursor

Stale Bugbot comment from a previous run.

Folding `..` before the config walk also moved the reported vendor directory onto the folded path, so path-scoped scan and rollback, which strip the given cwd as a prefix, saw the vendored crates as outside the project. The folded path now only picks the config files; a directory under it is reported back under the cwd as given. Refs #338 Assisted-by: Claude Code:claude-opus-5-5

Mikola Lysenko (mikolalysenko) · 2026-10-12T10:26:04Z

BugBot review

Generated by Claude Code

cursor

Stale Bugbot comment from a previous run.

The new prefix test compared crawler output with a path that still held `..`, but on Windows `std::path::absolute` folds it, so the crawler (and path-scoped commands) report the folded path there. The test now builds its expectation with `absolute` too. Refs #338 Assisted-by: Claude Code:claude-opus-5-5

Mikola Lysenko (mikolalysenko) · 2026-10-12T10:36:22Z

BugBot review

Generated by Claude Code

cursor

Stale Bugbot comment from a previous run.

Mikola Lysenko (mikolalysenko) · 2026-10-12T11:11:06Z

Final review brief (update for `747c561`)

The earlier brief covered everything up to 9083269. This one reviews the three commits since then.

What it does: Before walking the ancestor .cargo/config* files, the cwd is now folded lexically to remove .. and .. So --cwd a/b/../proj reads a/proj's configs and no longer picks up the sibling a/b/.cargo; this was open question 1 in the earlier brief. A directory source that resolves under the folded cwd is still reported under the cwd exactly as the user gave it, so path-scoped commands can strip it as a prefix. The rest of the PR is unchanged from the earlier brief.

Risk: low. The change is small and pure: +30/-4 in one function plus two tests. .. above the root is a no-op, because PathBuf::pop stops at /, C:\ and UNC roots.

Look here:

cargo_crawler.rs#L305-L318:`` fold_parent_dirs
cargo_crawler.rs#L320-L357:``` the split between given and `folded`, and the `strip_prefix(&folded)` → `given.join(rel)` remap
crawler_cargo_e2e.rs#L1030-L1075:```` the two new regression tests

Verified:

Read the delta 9083269..747c561.
Local runs: crawler_cargo_e2e 41/41, cargo_crawler unit tests 35/35, source_replacement tests 9/9.
Both new tests fail without the fix. parent_dir_in_cwd_does_not_read_a_sibling_config fails against 9083269, and directory_source_keeps_the_given_cwd_prefix fails against ae078c1.
CI ci-ok and clippy are green on 747c561. Bugbot was requested on that head.
No unresolved threads, and CHANGELOG.md is untouched.

Changes I made: none.

Open questions (non-blocking):

The fold is lexical. Cargo walks the physical cwd, so for link/../app, where link is a symlink, Cargo reads the configs above the link's target instead. The code before this PR also got this wrong, so it is not a regression.
On Windows, std::path::absolute already folds ... directory_source_keeps_the_given_cwd_prefix therefore only guards the remap on Unix and macOS.

Auto-merge is armed, so approving sends this straight to the merge queue.

Generated by Claude Code

The hosted refusal for a `cargo vendor` project only read the project's own `.cargo/config*`. A directory source set in a monorepo root's config or in `$CARGO_HOME` was missed, and a `cargo local-registry` was never recognised, so `scan --mode hosted` still pinned the Socket registry and every `cargo build --frozen --offline` broke. The hosted scan now resolves the source replacement from every config cargo reads for a build there, read-only, and passes it into the cargo rewriter. A directory source or local registry from any of them skips every cargo dependency with `redirect_cargo_directory_source`. Agent mode keeps crawling the registry cache for a local registry, which cargo unpacks there. The process-level resolve now lives in `source_replacement` so the crawler and the hosted scan share it. Refs #455 Assisted-by: Claude Code:claude-opus-5-5

Mikola Lysenko (mikolalysenko) · 2026-10-12T16:00:56Z

BugBot review

Generated by Claude Code

cursor

Stale Bugbot comment from a previous run.

The format models are pure by rule (formats::architecture_tests), but the last commit put the file-reading resolve into formats/cargo/source_replacement.rs. It now lives in the cargo crawler as `CargoCrawler::crates_io_source_for`, which the hosted scan calls; the format module keeps only the pure chain resolution. Refs #455 Assisted-by: Claude Code:claude-opus-5-5

Mikola Lysenko (mikolalysenko) · 2026-10-12T16:23:36Z

BugBot review

Generated by Claude Code

cursor

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit d81161b. Configure here.}

Mikola Lysenko (mikolalysenko) · 2026-10-12T16:50:02Z

Final reviewer: disarmed auto-merge. My earlier brief covered 747c561; two non-merge commits have landed since (4032716 "Refuse hosted cargo pins on outer offline sources", d81161b "Keep cargo config reads out of formats"). I'll write a fresh brief for d81161b on a later run once the head has settled, then re-arm.

Generated by Claude Code

Main's #1420 added a restored field to the e2e Shape struct; the declared-cargo-floor test this branch adds predates it and no longer compiled after merging main. Co-Authored-By: Claude <noreply@anthropic.com>

Start fix for #338, #455

386b86c

Assisted-by: Claude Code:claude-opus-5-5

Claude (claude) added 5 commits October 12, 2026 06:42

Mikola Lysenko (mikolalysenko) marked this pull request as ready for review October 12, 2026 07:33

cursor Bot reviewed Oct 12, 2026

View reviewed changes

Mikola Lysenko (mikolalysenko) mentioned this pull request Oct 12, 2026

Swarm health manager log #1382

Open

Mikola Lysenko (mikolalysenko) added the Ready for review Agent-verified: mergeable, CI green, Bugbot clean — awaiting human review label Oct 12, 2026

Mikola Lysenko (mikolalysenko) enabled auto-merge October 12, 2026 10:03

cursor Bot reviewed Oct 12, 2026

View reviewed changes

Comment thread crates/socket-patch-core/src/crawlers/cargo_crawler.rs Outdated

cursor Bot reviewed Oct 12, 2026

View reviewed changes

Comment thread crates/socket-patch-core/tests/crawler_cargo_e2e.rs Outdated

cursor Bot reviewed Oct 12, 2026

View reviewed changes

Mikola Lysenko (mikolalysenko) disabled auto-merge October 12, 2026 10:50

Mikola Lysenko (mikolalysenko) enabled auto-merge October 12, 2026 11:11

Tanmay Singla (Tanmay182003) reviewed Oct 12, 2026

View reviewed changes

Comment thread crates/socket-patch-core/src/patch/redirect/mod.rs Outdated

cursor Bot reviewed Oct 12, 2026

View reviewed changes

Tanmay Singla (Tanmay182003) reviewed Oct 12, 2026

View reviewed changes

Comment thread crates/socket-patch-core/src/formats/cargo/source_replacement.rs Outdated

cursor Bot reviewed Oct 12, 2026

View reviewed changes

Mikola Lysenko (mikolalysenko) mentioned this pull request Oct 12, 2026

Annotate failing tests in the coverage job #1438

Merged

Mikola Lysenko (mikolalysenko) disabled auto-merge October 12, 2026 16:49

Tanmay Singla (Tanmay182003) approved these changes Oct 12, 2026

View reviewed changes

Mikola Lysenko (mikolalysenko) mentioned this pull request Oct 12, 2026

CI flake: coverage job fails intermittently in vendor / in_process_edge_cases #1416

Closed

Mikola Lysenko (mikolalysenko) added this pull request to the merge queue Oct 12, 2026

Merged via the queue into main with commit 7236190 Oct 12, 2026
54 checks passed

Mikola Lysenko (mikolalysenko) deleted the agent/fix-cargo-source-replacement branch October 12, 2026 17:58

Mikola Lysenko (mikolalysenko) mentioned this pull request Oct 12, 2026

Fix hosted/vendored cargo floor reads (#653, #651) #1429

Open

Conversation

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

LLM Description written by Claude Code:claude-opus-5-5

Summary

Per-issue checklist

Local evidence (on 30aa879; 9083269 adds 3 fixture wirings, CI green)

/code-review high: findings not fixed here

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

cursor Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

cursor Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Final review brief

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

cursor Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

cursor Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

cursor Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Final review brief (update for 747c561)

Uh oh!

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

cursor Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

cursor Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Mikola Lysenko (mikolalysenko) commented Oct 12, 2026 •

edited

Loading

Local evidence (on `30aa879`; `9083269` adds 3 fixture wirings, CI green)

cursor Bot left a comment •

edited

Loading

cursor Bot left a comment •

edited

Loading

cursor Bot left a comment •

edited

Loading

cursor Bot left a comment •

edited

Loading

cursor Bot left a comment •

edited

Loading

Final review brief (update for `747c561`)

cursor Bot left a comment •

edited

Loading