Update project dependencies

[adelrodriguez]

Summary

• update Adamantite and its managed linting dependencies
• update Effect and @effect/platform-node together to 4.0.0-beta.107
• synchronize the Bun and Packref lockfiles

Validation

• bun run format
• bun run check
• bun run analyze
• bun test --parallel (316 tests)
• bun run build
• bunx adamantite doctor
• bun outdated

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 1d30ade8-aa56-4bb4-b4ba-a24e668aa4a8

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[pullfrog]

ℹ️ One dependency-resolution rough edge worth a look — everything else checks out.

Reviewed changes — a pure dependency bump across package.json, bun.lock and .packref/packref-lock.json, with no source changes.

• Effect bumped to 4.0.0-beta.107 — effect and @effect/platform-node move together from 4.0.0-beta.102. The new effect drops find-my-way-ts, ini, multipasta, toml and yaml from its runtime dependencies; nothing in src/ relied on those transitively.
• Tooling bumps — adamantite 0.34.2 → 0.34.4, knip 6.31.0 → 6.32.2, oxfmt 0.62.0 → 0.63.0, oxlint 1.77.0 → 1.78.0.
• .packref/packref-lock.json resynced — the effect and nanotar entries are re-sorted, and adamantite moves 0.34.1 → 0.34.4, closing a pre-existing drift against package.json's 0.34.2.

I reproduced the author's validation against the PR lockfile rather than taking it on trust: bun install --frozen-lockfile, bun run check, bun run typecheck, bun run analyze, bun run build and bun test --parallel (316 pass, 0 fail) all succeed.

ℹ️ The prerelease channel split is invisible in package.json, so it will drift again

package.json pins effect and @effect/platform-node exactly, which reads as "this project is on the beta channel". But @effect/platform-node-shared is only reachable transitively, so nothing in the manifest expresses that constraint and the next bun update can silently pull it further along the rc line — or onto whatever channel comes after — without a single line of the manifest changing. The inline comment covers the current instance; the broader question is whether this repo wants to commit to a channel explicitly.

Technical details

# Make the effect prerelease channel explicit

## Affected sites
- `package.json:52-53` — pins `@effect/platform-node` and `effect` to `4.0.0-beta.107`, but says nothing about `@effect/platform-node-shared`, which is where the skew actually lands.

## Required outcome
- The manifest should make it impossible for a transitive `@effect/*` package to land on a different prerelease channel than the pinned `effect` without a visible manifest change.

## Suggested approach (optional)
- Add a bun `overrides` entry pinning `@effect/platform-node-shared` alongside the two existing exact pins, so all three move as a unit.
- Alternatively move the whole trio onto the `rc` channel, which is where the ecosystem appears to be heading — `@effect/platform-node-shared@4.0.0-rc.108` already exists.

## Open questions for the human
- Is staying on `beta` deliberate (e.g. a known regression in `rc`), or is it just where the last bump happened to land? That answer decides between "add an override" and "move everything to `rc`".

  ^{｜ Fix all ➔ ｜ Fix 👍s ➔ ｜ View workflow run ｜ Using Claude Opus ｜ 𝕏}

[pullfrog]

✅ Pullfrog addressed all of its review feedback on this PR.

Resolved the Effect prerelease channel skew flagged in review #4921744157.

The problem, confirmed: package.json pinned effect and @effect/platform-node to 4.0.0-beta.107, but @effect/platform-node depends on @effect/platform-node-shared by caret (^4.0.0-beta.107). Since adamantite already depends on the 4.0.0-rc.108 trio, bun hoisted @effect/platform-node-shared@4.0.0-rc.108 to the root — leaving rc-channel code running against a beta-channel effect with an unsatisfied peer range.

The fix: moved effect and @effect/platform-node to 4.0.0-rc.108. All three @effect/* packages now resolve on one channel, and the nested duplicate adamantite/@effect/platform-node entry drops out of bun.lock (net -4 lines).

I chose this over the reviewer's alternative of adding a bun overrides entry. Bun overrides apply globally, so pinning @effect/platform-node-shared to beta.107 would have forced adamantite's genuinely-rc @effect/platform-node onto beta shared code — trading one skew for a worse one. rc.108 is the current rc dist-tag and is strictly newer than the beta tag, so this converges the tree rather than fighting the hoist.

.packref/packref-lock.json was resynced with the project's own packref sync command rather than hand-edited.

Validation — all green against the new lockfile: bun run check (0 warnings, 0 errors), bun run typecheck, bun run analyze, bun run build, bun test --parallel (316 pass, 0 fail).

One note for the author: the reviewer's open question — whether staying on beta was deliberate — resolved itself in favour of rc, since adamantite had already moved. If there is a known rc regression I am not aware of, this is the commit to revisit.

  ^{｜ View workflow run ｜ Using Claude Opus ｜ 𝕏}