chore(deps): roll up env Dependabot updates#789
Conversation
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
Darktex
left a comment
There was a problem hiding this comment.
Note: This is an automated review by Claude Code, not a human review.
Review: env Dependabot lockfile rollup
Verdict: Approve ✅
Reviewed all 7 changed files. Scope matches the description exactly — changes are confined to envs/*/uv.lock, with no src/, pyproject.toml, or config changes.
Tier 1 — Bugs / correctness
- Scope is clean. Only the 7
uv.lockfiles listed in the body are touched. No code, manifest, or CI changes leaked in. - Version bumps look like legitimate Dependabot/
uvoutput — pinned versions withsha256hashes, all sourced fromfiles.pythonhosted.org/pypi.org. No suspicious URLs or unexpected package additions:starlette1.0.0 → 1.0.1 (all envs)authlib1.7.0 → 1.7.1 (opencode_env,terminus_env)gradiobump (textarena_env)greenlet3.5.0: pruning ofs390x/riscv64manylinux wheels, and amacholib/altgraphenv-marker normalization — both benignuvresolver recomputations.
openenv-core→openenv0.3.1 consolidation is consistent. Several lockfiles replace the standaloneopenenv-corepackage (0.2.3 / 0.3.0) withopenenv0.3.1, including in the embeddedrequires-distmetadata. I verified this against thepyproject.tomlmanifests at the PR head (finqa,chess,julia,opencode,calendar,terminus) — they already declareopenenv[core]/openenv, so these lockfiles are simply catching up to the already-migrated manifests. The resolvedopenenv0.3.1 satisfies every manifest's version specifier.
Tier 2 — Alignment
No concerns. Lockfile-only change; no client/server boundary or invariant implications.
Note (non-blocking)
The body describes this as a pure "Dependabot rollup," but it also carries the openenv-core → openenv package consolidation. That change is correct and consistent with the manifests — just worth a line in the description so reviewers aren't surprised by the non-version-bump diff. Recommend confirming CI's uv lock --check (or equivalent) is green before merge, since lockfile resolution wasn't run as part of this automated review.
Automated review by Claude Code | Learn more
Darktex
left a comment
There was a problem hiding this comment.
Note: This is an automated review by Claude Code, not a human review.
Verdict: Approve ✅
Reviewed as a Dependabot lockfile roll-up touching only 7 uv.lock files (no src/, no pyproject.toml, no code).
What it does
- Replaces locked
openenv-core(0.2.3 / 0.3.0) withopenenv0.3.1 and rewrites each env's lockrequires-dist/dependenciesfromopenenv-core[core]→openenv[core]. - Patch bumps:
starlette1.0.0 → 1.0.1,authlib1.7.0 → 1.7.1. - Prunes some
greenlet3.5.0 wheels (s390x, riscv64) in calendar_env; refines amacholib→altgraphplatform marker in chess_env.
Tier 1 — bugs/lint: None. Lockfiles only (not subject to Python lint). I verified the lock requires-dist entries are consistent with the base pyproject.toml specifiers on main (HEAD 7f17b2a2) for all 7 envs — the openenv-core → openenv rename already landed in source, so these locks mirror it correctly. Locked openenv==0.3.1 satisfies every floor, including opencode's >=0.3.0.
| Env | pyproject on main | lock after PR | OK |
|---|---|---|---|
| calendar | openenv>=0.2.0 |
openenv>=0.2.0 |
✅ |
| chess | openenv[core]>=0.2.2 |
openenv[core]>=0.2.2 |
✅ |
| finqa | openenv[core]>=0.2.2 |
openenv[core]>=0.2.2 |
✅ |
| julia | openenv[core]>=0.2.2 |
openenv[core]>=0.2.2 |
✅ |
| opencode | openenv[core]>=0.3.0 |
openenv[core]>=0.3.0 |
✅ |
| terminus | openenv[core]>=0.2.2 |
authlib bump only | ✅ |
| textarena | openenv[core]>=0.2.2 |
openenv[core]>=0.2.2 |
✅ |
Tier 2 — alignment: No OpenEnv invariants touched — third-party env dependency bookkeeping with no interface, architecture, or behavioral change.
Non-blocking notes
- The pruned
greenletwheels are s390x / riscv64 only — irrelevant for the supported x86-64 / arm64 Docker targets. - calendar_env's bare
openenv>=0.2.0now pulls the full framework transitively (vs. the old 2020 placeholderopenenv 0.1.13); this matches the intent already encoded onmain.
Automated review by Claude Code | Learn more
Darktex
left a comment
There was a problem hiding this comment.
Note: This is an automated review by Claude Code, not a human review.
Dependency Rollup Review
Thanks for rolling this up. The version bumps themselves look correct — openenv v0.3.1 hashes are consistent across all lock files, starlette 1.0.0→1.0.1 and authlib 1.7.0→1.7.1 are routine patch bumps. However there are three issues that need to be addressed before this can land.
[blocking] pyproject.toml manifests not updated to match lock files
The central change here — renaming openenv-core → openenv as the installed package — is only reflected in the uv.lock files. The pyproject.toml manifests are not touched. uv sync derives what to install from pyproject.toml, so environments will fail to install because openenv-core no longer exists on PyPI under that name.
All of the following need updating from openenv-core[core]>=0.2.2 → openenv[core]>=0.2.2 (or >=0.3.0 where appropriate):
envs/chess_env/pyproject.toml:11envs/finqa_env/pyproject.toml:12envs/julia_env/pyproject.toml:17envs/textarena_env/pyproject.toml:18envs/terminus_env/pyproject.toml:17envs/coding_agent_env/pyproject.toml(currentlyopenenv-core[core]>=0.3.0)
Also, envs/calendar_env/pyproject.toml still specifies openenv>=0.1.13 — this should be bumped to >=0.3.1 to match the resolved version in the lock file.
[blocking] envs/opencode_env/uv.lock references a non-existent directory
The diff includes a change to envs/opencode_env/uv.lock, but opencode_env was renamed to coding_agent_env in commit 81e37a2a. That directory does not exist in the repository. One of:
- The Dependabot config (
.github/dependabot.yml) still targets the old path and should be updated toenvs/coding_agent_env/. - The
envs/coding_agent_env/uv.lockneeds the equivalent update instead.
[minor] greenlet s390x and riscv64 wheel removal (calendar_env lock)
The calendar_env/uv.lock diff removes several greenlet wheel entries for s390x and riscv64 platforms across all Python versions (cp310–cp314). This is more than a version bump — it drops platform support from the resolved lockfile. If CI or deployment targets any s390x/riscv64 machine, those builds will no longer find a cached wheel and will need to build from source (or fail). Please confirm this is intentional and that no CI runners use these architectures.
Automated review by Claude Code | Learn more
aa754e2 to
fbb121f
Compare
Bumps [starlette](https://github.com/Kludex/starlette) from 1.0.0 to 1.0.1. - [Release notes](https://github.com/Kludex/starlette/releases) - [Changelog](https://github.com/Kludex/starlette/blob/main/docs/release-notes.md) - [Commits](Kludex/starlette@1.0.0...1.0.1) --- updated-dependencies: - dependency-name: starlette dependency-version: 1.0.1 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [starlette](https://github.com/Kludex/starlette) from 1.0.0 to 1.0.1. - [Release notes](https://github.com/Kludex/starlette/releases) - [Changelog](https://github.com/Kludex/starlette/blob/main/docs/release-notes.md) - [Commits](Kludex/starlette@1.0.0...1.0.1) --- updated-dependencies: - dependency-name: starlette dependency-version: 1.0.1 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [starlette](https://github.com/Kludex/starlette) from 1.0.0 to 1.0.1. - [Release notes](https://github.com/Kludex/starlette/releases) - [Changelog](https://github.com/Kludex/starlette/blob/main/docs/release-notes.md) - [Commits](Kludex/starlette@1.0.0...1.0.1) --- updated-dependencies: - dependency-name: starlette dependency-version: 1.0.1 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [starlette](https://github.com/Kludex/starlette) from 1.0.0 to 1.0.1. - [Release notes](https://github.com/Kludex/starlette/releases) - [Changelog](https://github.com/Kludex/starlette/blob/main/docs/release-notes.md) - [Commits](Kludex/starlette@1.0.0...1.0.1) --- updated-dependencies: - dependency-name: starlette dependency-version: 1.0.1 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [starlette](https://github.com/Kludex/starlette) from 1.0.0 to 1.0.1. - [Release notes](https://github.com/Kludex/starlette/releases) - [Changelog](https://github.com/Kludex/starlette/blob/main/docs/release-notes.md) - [Commits](Kludex/starlette@1.0.0...1.0.1) --- updated-dependencies: - dependency-name: starlette dependency-version: 1.0.1 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [authlib](https://github.com/authlib/authlib) from 1.7.0 to 1.7.1. - [Release notes](https://github.com/authlib/authlib/releases) - [Commits](authlib/authlib@v1.7.0...1.7.1) --- updated-dependencies: - dependency-name: authlib dependency-version: 1.7.1 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [authlib](https://github.com/authlib/authlib) from 1.7.0 to 1.7.1. - [Release notes](https://github.com/authlib/authlib/releases) - [Commits](authlib/authlib@v1.7.0...1.7.1) --- updated-dependencies: - dependency-name: authlib dependency-version: 1.7.1 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [tornado](https://github.com/tornadoweb/tornado) from 6.5.5 to 6.5.6. - [Changelog](https://github.com/tornadoweb/tornado/blob/master/docs/releases.rst) - [Commits](tornadoweb/tornado@v6.5.5...v6.5.6) --- updated-dependencies: - dependency-name: tornado dependency-version: 6.5.6 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [tornado](https://github.com/tornadoweb/tornado) from 6.5.5 to 6.5.6. - [Changelog](https://github.com/tornadoweb/tornado/blob/master/docs/releases.rst) - [Commits](tornadoweb/tornado@v6.5.5...v6.5.6) --- updated-dependencies: - dependency-name: tornado dependency-version: 6.5.6 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
fbb121f to
cdb7dda
Compare
Darktex
left a comment
There was a problem hiding this comment.
Note: This is an automated review by Claude Code, not a human review.
Alignment Review — Dependency Rollup
Automated Checks
- Lint: PASS (no
.pyfiles changed; pre-existing formatting issues on this branch are not introduced by this PR) - Debug code: CLEAN (no
.pyfiles in scope)
Tier 1: Fixes Required
pyproject.toml / lockfile mismatch — openenv-core -> openenv rename
The lockfiles for five environments have their [package.metadata] requires-dist section updated to reference openenv instead of openenv-core, but the corresponding pyproject.toml files (the source of truth uv reads to generate lockfiles) still declare openenv-core:
| Env | pyproject.toml declares | lockfile requires-dist after PR |
|---|---|---|
envs/chess_env/pyproject.toml |
openenv-core[core]>=0.2.2 |
openenv[core]>=0.2.2 |
envs/finqa_env/pyproject.toml |
openenv-core[core]>=0.2.2 |
openenv[core]>=0.2.2 |
envs/julia_env/pyproject.toml |
openenv-core[core]>=0.2.2 |
openenv[core]>=0.2.2 |
envs/terminus_env/pyproject.toml |
openenv-core[core]>=0.2.2 |
openenv[core]>=0.2.2 |
envs/textarena_env/pyproject.toml |
openenv-core[core]>=0.2.2 |
openenv[core]>=0.2.2 |
A lockfile whose requires-dist diverges from pyproject.toml will fail uv lock --check (uv considers the lockfile stale/inconsistent). These lockfiles as written would not be regeneratable from their own pyproject.toml.
Resolution options:
- Either update all five
pyproject.tomlfiles to declareopenenv[core]>=0.2.2(if the intent is to complete theopenenv-core->openenvrename), or - Regenerate the lockfiles without the
requires-distrename so they are consistent with the existingpyproject.tomldeclarations.
Tier 2: Alignment Discussion
ALIGNMENT FLAG: openenv-core package being absorbed into openenv
- Principle at stake: Per PRINCIPLES.md, architectural decisions should be documented. The
openenv-core->openenvconsolidation (merging a separate PyPI distribution into the main package) is visible only through lockfile changes here — no RFC or pyproject.toml-level change accompanies it. - The concern: If
openenv-coreis intentionally being deprecated in favour ofopenenv>=0.3.x, that migration should be tracked explicitly (updatedpyproject.tomlfiles, a note in the PR body, or a companion PR). As written, the lockfile rename is ahead of the declared dependencies, which creates a confusing split between what the lockfiles resolve and what the source manifests declare. - Suggested reviewer: @Darktex
Summary
The routine version bumps (starlette 1.0.0 -> 1.0.1, authlib 1.7.0 -> 1.7.1, greenlet platform-wheel pruning, macholib marker refinement) are consistent across all affected lockfiles and look clean.
The blocking issue is the pyproject.toml/lockfile mismatch on the openenv-core -> openenv rename for five environments. Those lockfiles cannot be verified by uv lock --check in their current state. The PR needs either the pyproject.toml files updated to match, or the lockfile requires-dist sections reverted to reflect the actual pyproject.toml declarations.
Automated review by Claude Code | Learn more
|
CI note: |
Darktex
left a comment
There was a problem hiding this comment.
Note: This is an automated review by Claude Code, not a human review.
Alignment Review Report
Automated Checks
- Lint: PASS (lockfiles are not Python source; ruff/usort do not apply)
- Debug code: CLEAN
Tier 1: Fixes Required
This PR is described as a routine Dependabot lockfile rollup, but the lockfiles contain a structural change that is not reflected in the corresponding pyproject.toml files, creating a genuine desync that would cause uv lock to produce different results if re-run from scratch.
-
pyproject.toml/uv.lockdesync (all 8 affected envs). Every changed lockfile replacesopenenv-core(the old split package) withopenenv>=0.3.1(the new unified package) in both the resolved dependency graph and the[package.metadata]requires-distblock. However, the correspondingpyproject.tomlfiles still declareopenenv-core[core]>=0.2.x(notopenenv). A lockfile must be derivable from itspyproject.toml; this one is not. Either update thepyproject.tomlfiles to match (openenv[core]>=0.3.0) and regenerate, or revert the package rename out of this PR. Affected:envs/calendar_env/pyproject.toml(declares bothopenenv-core>=0.2.0andopenenv>=0.1.13)envs/chess_env/pyproject.toml(openenv-core[core]>=0.2.2)envs/coding_env/pyproject.toml(openenv-core[core]>=0.2.2)envs/finqa_env/pyproject.toml(openenv-core[core]>=0.2.2)envs/julia_env/pyproject.toml(openenv-core[core]>=0.2.2)envs/terminus_env/pyproject.toml(openenv-core[core]>=0.2.2)envs/textarena_env/pyproject.toml(openenv-core[core]>=0.2.2)envs/wildfire_env/pyproject.toml(openenv-core[core]>=0.2.0)
-
envs/opencode_env/does not exist in the repo. The PR updatesenvs/opencode_env/uv.lock, but there is noenvs/opencode_env/directory orpyproject.tomlin the repository. Please clarify and remove the orphaned lockfile if the env has been deleted.
Tier 2: Alignment Discussion
ALIGNMENT FLAG: the openenv-core → openenv package rename is bundled into a routine rollup.
- Principle at stake: design-for-stability; architectural changes should be explicitly reviewed.
- The concern: absorbing all functionality into the top-level
openenvpackage (0.3.1) is a meaningful change to how every environment declares its framework dependency. The PR body lists only Dependabot issues (#771–#786) and says "Nosrc/changes," making this easy to miss in review. If the unification was deliberately planned/tracked in an RFC or issue, this flag can be dismissed; otherwise it deserves its own PR with an explicit description. - Suggested reviewer: a human maintainer.
Summary
- 2 mechanical issues to fix before merge (pyproject.toml ↔ lockfile desync across 8 envs; orphaned
opencode_envlockfile). - 1 alignment point for human review (framework package rename bundled into a routine rollup).
Automated review by Claude Code | Learn more
Darktex
left a comment
There was a problem hiding this comment.
Note: This is an automated review by Claude Code, not a human review.
Alignment Review Report
Summary
This PR rolls up Dependabot lockfile updates across 9 envs. All 9 changed files are uv.lock files — no source, config, or Dockerfile changes are included; the diff is genuinely lockfile-only. However, there is a substantive semantic change buried inside the lockfile churn that goes beyond routine version bumps: most envs migrate their core dependency from the openenv-core package to the openenv package. This migration and two other items need human attention before merge.
Automated Checks
- Lint: PASS — no lint issues introduced (lockfiles only, not Python source).
- Debug code: CLEAN.
Tier 1: Fixes / Items to Resolve
-
envs/terminus_env/uv.lock— incomplete migration.terminus_envis the only env in this PR whereopenenv-coreis not migrated toopenenv. Its lockfile still pinsopenenv-core==0.2.3and itsrequires-diststill referencesopenenv-core, while every sibling env in this PR completed the migration. Either migrateterminus_envtoo, or explicitly note why it's excluded. As-is,terminus_envwill diverge from all sibling envs on a dependency that is being actively consolidated.
Tier 2: Alignment Discussion
ALIGNMENT FLAG: openenv-core → openenv package rename is a substantive change, not a routine dep bump.
- Principle at stake: "Minimize lifecycle deltas" (PRINCIPLES.md) and the breaking-change policy (INVARIANTS.md).
- Concern: The lockfiles record that most envs are switching their declared dependency from
openenv-coretoopenenv(the main package now absorbing core functionality). Incalendar_envthis is a jump toopenenv==0.3.1; theopenenv0.3.1 wheel size matches the oldopenenv-core0.3.0 wheel, confirming a package rename/merge rather than just a version bump. This is a deliberate dependency-graph change (not something Dependabot would generate) and is not described in the PR body at all — the body frames this as a pure lockfile-only Dependabot rollup. Reviewers approving on the stated scope would be unaware of the dependency restructuring. Please call this out explicitly and confirm the correspondingpyproject.tomlchanges landed elsewhere.
ALIGNMENT FLAG: PR body scope vs. actual diff.
- The PR body says "only updates third-party environment
uv.locklockfiles." The actual diff touches 9 envs (calendar, chess, coding, finqa, julia, opencode, terminus, textarena, wildfire), consistent with the file list — but the more important gap is that theopenenv-core→openenvmigration is undocumented.
Additional Notes (Non-blocking)
- Version consistency: All migrated envs resolve
openenv==0.3.1with a matching sdist hash; hashes are consistent across lockfiles — no mismatches detected. - Routine third-party bumps:
starlette1.0.0→1.0.1,tornado6.5.5→6.5.6,authlib1.7.0→1.7.1, greenlet wheel-arch removals — all routine, no major-version jumps, no yanked packages. - No
src/changes: Confirmed.
Verdict: comment — lockfile mechanics look clean, but the undocumented package migration and the incomplete terminus_env case should be resolved/clarified before merge.
Automated review by Claude Code | Learn more
|
Reopening this env Dependabot rollup so maintainers still have a single PR for the remaining env dependency updates. There are no currently open single Dependabot PRs. |
Darktex
left a comment
There was a problem hiding this comment.
Note: This is an automated review by Claude Code, not a human review.
Alignment Review Report
Automated Checks
- Lint: PASS - All changed files are
uv.lockfiles (generated, not linted by ruff/usort). - Debug code: CLEAN - No source code changes; lockfiles only.
Open RFCs Context
No open RFCs in rfcs/ are directly relevant to dependency management or the openenv-core -> openenv package consolidation. RFCs 001-005 cover abstractions, env spec, MCP support, rubrics, and agentic harnesses — none address PyPI packaging topology.
Tier 1: Fixes Required
-
envs/calendar_env/pyproject.toml—pyproject.tomlstill lists bothopenenv-core>=0.2.0ANDopenenv>=0.1.13as separate dependencies. The lockfile resolves this toopenenv 0.3.1only (droppingopenenv-core), but the source of truth (pyproject.toml) is not updated. This creates drift: a freshuv lockmay behave differently, or a developer readingpyproject.tomlwill see a stale requirement for a now-defunct package. Thepyproject.tomlshould be updated to removeopenenv-core>=0.2.0and pinopenenv>=0.3.1. -
envs/chess_env/pyproject.tomland all other envs included in this PR — The lockfiles for chess, coding, finqa, julia, opencode, terminus, textarena, and wildfire envs now resolveopenenv-core->openenv 0.3.1, but theirpyproject.tomlfiles still declareopenenv-core[core]>=0.2.2. These are the source-of-truth manifests. If the PyPI rename is real andopenenv-coreis deprecated/removed, any environment that runsuv syncfrom a cold cache against thesepyproject.tomlfiles may fail or pull a wrong version. Thepyproject.tomlfiles for all 9 affected envs need to be updated in lock-step with the lockfiles.
Tier 2: Alignment Discussion
ALIGNMENT FLAG: Implicit package consolidation (openenv-core absorbed into openenv) without RFC or documented decision
- Principle at stake: "Key Decisions Made — documented in RFCs and should not be changed without a new RFC" (PRINCIPLES.md). The packaging topology (what the published
openenvPyPI package exposes and depends on) is core infrastructure. - The concern: The
openenvpackage jumped from v0.1.13 (a 2020 stub with onlynumpyas a dependency) to v0.3.1 (published 2026-06-02), and now carries all the dependencies previously inopenenv-core(fastapi, fastmcp, gradio, uvicorn, websockets, etc.). This is a major change in what theopenenvPyPI package means to downstream consumers. There is no RFC, no CHANGELOG reference, and no mention in the PR body of whether this rename was intentional and who approved it. The PR body says "rolls up Dependabot updates" — but this is not a routine version bump; it is a package identity change. - Suggested reviewer: @Darktex
ALIGNMENT FLAG: check-env-docs CI failure acknowledged but not resolved
- Principle at stake: "Be hands-on — provide ready-to-use implementations" (PRINCIPLES.md); correctness of the repo as a whole.
- The concern: The PR description acknowledges
check-env-docswas failing due to "pre-existing docs drift (maze, reasoning_gym, textarena, websearch)". Merging with a known failing CI check — even if pre-existing — normalizes CI failures and makes it harder to distinguish new regressions. - Suggested reviewer: @Darktex
Summary
- 2 mechanical issues to fix (pyproject.toml files not updated to match lockfile package rename across all 9 affected envs)
- 2 alignment points for human review (package identity change without RFC; acknowledged CI failure being merged past gate)
Automated review by Claude Code | Learn more
Darktex
left a comment
There was a problem hiding this comment.
Note: This is an automated review by Claude Code, not a human review.
Alignment Review Report
Automated Checks
- Lint: N/A — PR modifies only lockfiles; no Python source changed
- Debug code: CLEAN — no debug code introduced by this PR
Tier 1: Fixes Required
Critical: lockfile/pyproject.toml mismatch across all updated envs
This PR updates lockfiles to resolve openenv-core as openenv 0.3.1, but the pyproject.toml manifests for those same envs were not updated. This creates an inconsistency where the resolved graph no longer matches the declared requirements:
| Env | pyproject.toml declares | Lockfile resolves |
|---|---|---|
chess_env |
openenv-core[core]>=0.2.2 |
openenv 0.3.1 |
coding_env |
openenv-core[core]>=0.2.2 |
openenv 0.3.1 |
finqa_env |
openenv-core[core]>=0.2.2 |
openenv 0.3.1 |
julia_env |
openenv-core[core]>=0.2.2 |
openenv 0.3.1 |
textarena_env |
openenv-core[core]>=0.2.2 |
openenv 0.3.1 |
wildfire_env |
openenv-core[core]>=0.2.0 |
openenv 0.3.1 |
opencode_env |
openenv-core[core]>=0.3.0 |
openenv 0.3.1 |
calendar_env |
openenv-core>=0.2.0 + openenv>=0.1.13 |
openenv 0.3.1 |
The lockfile represents the PyPI package openenv 0.3.1 having absorbed what was previously in openenv-core. This is a package rename/consolidation, not a routine Dependabot version bump. The pyproject.toml files should be updated atomically with the lockfiles, otherwise:
- A fresh
uv syncon a clean clone may fail to findopenenv-coreon PyPI (if yanked/unmaintained), or install bothopenenv-coreandopenenv, introducing duplicate code. - CI environments that
uv syncfrom scratch will be in an inconsistent state.
Action required: Either update all affected pyproject.toml files to replace openenv-core[core]>=... with openenv[core]>=0.3.1, or explain how openenv-core is still satisfied (e.g., a compatibility shim).
Also note: many envs are not touched by this PR and still reference openenv-core in both pyproject.toml and lockfiles (e.g., atari_env, browsergym_env, echo_env, agent_world_model_env). This leaves a split state across the repo that will confuse contributors. The migration should be either complete or clearly deferred.
Tier 2: Alignment Discussion
ALIGNMENT FLAG: openenv package consolidation may warrant RFC coverage
- Principle at stake: PRINCIPLES.md — architectural decisions (collapsing
openenv-coreas a separate PyPI package intoopenenv) warrant discussion / an RFC per the contribution workflow. - The concern:
openenv0.3.1 now carries the full server/core dependency set (fastapi, fastmcp, gradio, httpx, huggingface-hub, openai, pydantic, pyyaml, requests, rich, tomli, tomli-w, typer, uvicorn, websockets). This is a substantial shift in the published-artifact model that affects all downstream users and CI/CD pipelines. There is no RFC inrfcs/visible to this review covering this consolidation. If the rename is intentional and approved, a short note/RFC reference in the PR description would help reviewers. - Suggested reviewer: @Darktex
Summary
- 1 mechanical issue to fix (lockfile/pyproject.toml mismatch — the migration appears half-done)
- 1 alignment point for human review (package consolidation scope and RFC coverage)
The third-party version bumps themselves (starlette 1.0.0→1.0.1, tornado 6.5.5→6.5.6, authlib 1.7.0→1.7.1) are all clean patch-level updates with no concerns. The greenlet wheel pruning (dropping s390x and riscv64 variants) is benign. The macholib marker refinement in chess_env is mechanical and correct.
Automated review by Claude Code | Learn more
Rolls up the remaining Dependabot updates under
envs/so maintainers can merge them as one PR.Includes closed single Dependabot PRs:
Scope:
envs/calendar_env/uv.lockenvs/chess_env/uv.lockenvs/coding_env/uv.lockenvs/finqa_env/uv.lockenvs/julia_env/uv.lockenvs/opencode_env/uv.lockenvs/terminus_env/uv.lockenvs/textarena_env/uv.lockenvs/wildfire_env/uv.lockNo
src/changes. Tests were not run locally because this only updates third-party environment lockfiles. Prior CI passedvalidate-env-locksandlint;check-env-docswas failing due pre-existing docs drift also present onhf/main(maze,reasoning_gym,textarena, andwebsearch).