The @demlik/tea catalog pin is 0.18.0 while 0.19.0 is published

[usirin]

In plain words

We pin tea at 0.18.0, but 0.19.0 is out and its built code is the same. Falling behind makes the next upgrade bigger. We would move the pin to 0.19.0 once the fabrika-cli 0.8.0 release merges.

What to do

Bump the root catalog pin for @demlik/tea from 0.18.0 to 0.19.0 in pnpm-workspace.yaml, refresh pnpm-lock.yaml, and keep the catalog comment naming the version the vitest-peer reason holds for.

The founder approved this bump on 2026-09-27. Land it only once release PR #9463 (fabrika-cli 0.8.0) has merged; check its state before opening the PR.

Triage note: read at origin/main: pnpm-workspace.yaml line 27 is '@demlik/tea': 0.18.0, and the comment above it (lines 24-26) names "0.18.0's optional vitest peer". Manifests listing @demlik/tea are packages/fabrika-cli, packages/tuval, packages/tuval-pi, packages/tuval-agy, packages/tuval-boot-proof, packages/tuval-claude, packages/tuval-cron, packages/tuval-notify, packages/tuval-shell, packages/tuval-worktree and apps/tuval. All resolve through the one catalog entry, so no manifest edit is expected.

Triage note: the filer reports dist/ is byte-identical between 0.18.0 and 0.19.0 and the peerDependencies match. I did not re-run that diff. If the lockfile refresh pulls anything beyond the tea entries, stop and report it.

Triage note: searched the board for tea pin work. #10133 (notice new tea releases) is a separate problem; #9787 and #9796 are closed and set the 0.18.0 pin. Nothing open owns this bump, so it stays standalone.

Acceptance criteria

• pnpm-workspace.yaml catalogs '@demlik/tea': 0.19.0, and the comment above it names 0.19.0 where it named 0.18.0.
• pnpm-lock.yaml resolves @demlik/tea@0.19.0 and has no @demlik/tea@0.18.0 entry left.
• No package manifest changes: every consumer still lists @demlik/tea as catalog:.
• The test suites of packages/fabrika-cli, packages/tuval and packages/tuval-pi pass, including fabrika-cli's tea-peer test.
• The PR merges only after release PR chore(main): release fabrika-cli 0.8.0 #9463 has merged. [evidence: state of https://github.com/chore(main): release fabrika-cli 0.8.0 #9463 at merge time]

---

Original report (verbatim)

Summary

The root catalog pins @demlik/tea at 0.18.0. tea 0.19.0 was published on 2026-09-26 and nobody has bumped the pin. The founder approved the bump on 2026-09-27, to land after the fabrika-cli 0.8.0 release PR (#9463) merges.

What I was doing

Preparing the fabrika-cli 0.8.0 release (#9463). While checking tea's version I saw 0.19.0 had been out for a day.

What I observed

• pnpm-workspace.yaml catalog: '@demlik/tea': 0.18.0 (set by fix(fabrika-cli)!: remove lane view and pin @demlik/tea 0.18.0, whose vitest peer admits the vitest adopters run #9784 for the vitest peer range, see @kampus/fabrika-cli pins @demlik/tea 0.12.0, whose vitest peer excludes the vitest adopters run #9771).
• tea 0.19.0 changelog: a ./agent usage fix, and ./effect graduates from experimental to stable.
• npm pack of 0.18.0 and 0.19.0 and a recursive diff: every file under dist/ is byte-identical. Only README.md (the ./effect stability label) and package.json (version, and a script rename verify-exports to verify:exports) differ.
• peerDependencies are identical between the two (same vitest: ^2 || ^3 || ^4 || ^5 range the pin exists for).
• Consumers: fabrika-cli imports only bare @demlik/tea. packages/tuval and packages/tuval-pi import @demlik/tea/effect (run, EffectRuntime, Stopped, StoreFailed); nothing in the workspace imports @demlik/tea/agent. Since the built code is unchanged, no consumer source change is expected.

Why it matters

Staying behind on a first-party dep makes the next bump bigger. Also the catalog comment should keep naming the version the vitest-peer reason holds for.

Pointers

• pnpm-workspace.yaml (catalog entry and its comment)
• pnpm-lock.yaml (@demlik/tea@0.18.0 entries)
• chore(main): release fabrika-cli 0.8.0 #9463, fix(fabrika-cli)!: remove lane view and pin @demlik/tea 0.18.0, whose vitest peer admits the vitest adopters run #9784, @kampus/fabrika-cli pins @demlik/tea 0.12.0, whose vitest peer excludes the vitest adopters run #9771
• Consumers: packages/fabrika-cli, packages/tuval, packages/tuval-pi, apps/tuval, and the other packages/tuval-* manifests that list @demlik/tea as catalog:

Suggested next step (non-binding)

After #9463 merges: bump the catalog entry to 0.19.0, update its comment to say 0.19.0, pnpm install to refresh the lockfile, and run the tests of fabrika-cli, tuval and tuval-pi.

---

_{Filed by an agent · session bc1bd9fd-00a1-474b-ada9-c223d0f09632 · branch main · 2026-09-28T05:43:08Z}

[usirin]

[usirin]

Opened #10134. It pins tea at 0.19.0 and touches only the tea rows in the lockfile. The fabrika-cli, tuval and tuval-pi suites pass, and so does the tea-peer test. No consumer source changed. Review the lockfile first: I applied the tea edits by hand because pnpm also dropped unrelated libc: lines, and --frozen-lockfile accepts the result.

[usirin]

{
"phases": [
{
"name": "pipeline",
"tasks": [
"issue"
]
}
],
"terminals": {
"complete": "complete",
"tripped": "tripped"
},
"tasks": {
"issue": {
"initial": "queued",
"maxRetries": 3,
"maxWaits": 3,
"maxLaps": 16,
"states": {
"queued": [
"WIP",
"BLOCKED",
"CLEARED",
"CANCELLED",
"LANDED"
],
"build": [
"DONE",
"BLOCKED",
"LAP",
"CLEARED",
"CANCELLED",
"LANDED"
],
"build:ui": [
"DONE",
"BLOCKED",
"LAP",
"CLEARED",
"CANCELLED",
"LANDED"
],
"review": [
"PASS",
"BLOCKED",
"LAP",
"FAIL",
"WIP",
"CLEARED",
"CANCELLED",
"LANDED"
],
"review:ui": [
"PASS",
"BLOCKED",
"LAP",
"FAIL",
"WIP",
"CLEARED",
"CANCELLED",
"LANDED"
],
"ship": [
"LAP",
"DONE",
"WIP",
"BLOCKED",
"FAIL",
"CLEARED",
"CANCELLED",
"LANDED"
],
"ship:queued": [
"LAP",
"DONE",
"BLOCKED",
"WIP",
"FAIL",
"CLEARED",
"CANCELLED",
"LANDED"
],
"blocked": [
"UNBLOCKED",
"CLEARED",
"CANCELLED",
"LANDED"
],
"human:cp-approval": [
"UNBLOCKED",
"CLEARED",
"CANCELLED",
"LANDED"
],
"human:queue-stall": [
"UNBLOCKED",
"CLEARED",
"CANCELLED",
"LANDED"
],
"human:machinery-stall": [
"UNBLOCKED",
"CLEARED",
"CANCELLED",
"LANDED"
],
"human:budget-spent": [
"UNBLOCKED",
"CLEARED",
"CANCELLED",
"LANDED"
],
"shipped": [
"CLEARED",
"CANCELLED",
"LANDED"
],
"diagnosed": [
"CLEARED",
"CANCELLED",
"LANDED"
],
"board:cancelled": [
"CLEARED"
],
"board:landed": [
"CLEARED"
]
}
}
}
}