Skip to content

ci(deps): update github-actions - #4

Open
renovate-bot-lohn[bot] wants to merge 1 commit into
mainfrom
renovate/github-actions
Open

ci(deps): update github-actions#4
renovate-bot-lohn[bot] wants to merge 1 commit into
mainfrom
renovate/github-actions

Conversation

@renovate-bot-lohn

@renovate-bot-lohn renovate-bot-lohn Bot commented Jul 12, 2026

Copy link
Copy Markdown

This PR contains the following updates:

Package Type Update Change Pending
actions/attest action minor v4.1.0v4.2.2
actions/checkout action patch v7.0.0v7.0.1
jdx/mise-action action patch v4.2.0v4.2.4 v4.2.5
marocchino/sticky-pull-request-comment action patch v3.0.4v3.0.5
pypa/gh-action-pypi-publish action patch v1.14.0v1.14.2

Release Notes

actions/attest (actions/attest)

v4.2.2

Compare Source

What's Changed

Full Changelog: actions/attest@v4.2.1...v4.2.2

v4.2.1

Compare Source

What's Changed

Full Changelog: actions/attest@v4.2.0...v4.2.1

v4.2.0

Compare Source

What's Changed

Full Changelog: actions/attest@v4.1.1...v4.2.0

v4.1.1

Compare Source

What's Changed

Full Changelog: actions/attest@v4.1.0...v4.1.1

actions/checkout (actions/checkout)

v7.0.1

Compare Source

jdx/mise-action (jdx/mise-action)

v4.2.4: : Reliable locking detection under forced color

Compare Source

A small patch release that fixes locking-support detection when workflows force colored output.

Fixed

Detect mise install --locked reliably under forced color (#​580 by @​scop)

When colored output was forced globally (for example via CLICOLOR_FORCE=1), ANSI escape codes in mise install --help prevented the action from matching --locked in the help text, so locking support was reported as unavailable even on versions of mise that supported it.

The help probe now runs with NO_COLOR=1 in its environment, which overrides CLICOLOR_FORCE and guarantees plain-text output for the feature detection — regardless of the surrounding workflow's color settings.

Full Changelog: jdx/mise-action@v4.2.3...v4.2.4

v4.2.3: : Restore mise PATH propagation

Compare Source

A patch release that restores mise's PATH propagation to subsequent workflow steps — without reintroducing the full-PATH snapshot behavior that v4.2.1 fixed.

Fixed

Export mise PATH entries to subsequent steps (#​575) by @​jdx

v4.2.1 stopped exporting the complete PATH returned by mise env --json into GITHUB_ENV, which correctly prevented snapshotting the runner's environment into subsequent steps. However, that also dropped mise-produced PATH entries — tool shims, [env] _.path directories, and similar — that workflows relied on after the setup step. See #​565.

The action now computes only the prefix that mise prepended to the existing PATH and forwards those directories individually through GITHUB_PATH. This preserves mise's configured ordering, composes cleanly with PATH changes from other actions, and never persists the runner's full PATH through GITHUB_ENV. The dotenv fallback path (used with older mise versions) also strips PATH= lines and re-derives additions from mise env --json.

A new export_path input (default true) lets workflows keep regular env exports while opting out of PATH changes:

- uses: jdx/mise-action@v4
  with:
    export_path: false # keep env vars, skip mise PATH additions

Full Changelog: jdx/mise-action@v4.2.2...v4.2.3

v4.2.2: : Zstd tar fallback for older runners

Compare Source

A small patch release that fixes archive selection on runners with an older tar and corrects a stale default in the README.

Fixed

Verify tar supports Zstd before picking .tar.zst (#​569 by @​JackMyers001

The action previously chose the .tar.zst mise archive whenever zstd --version succeeded, then extracted it with tar --zstd. On RHEL 8-compatible runners that ship zstd 1.4.4 alongside GNU tar 1.30, the --zstd option isn't recognized and installation failed.

Detection now runs both checks:

zstd --version
tar --zstd --version

If either fails, the action falls back to the .tar.gz archive. No configuration change is required — existing workflows on affected runners just start working again. Fixes #​568.

Documentation

  • Update the cache_key_prefix example in the README to reflect the current default of mise-v1 (previously documented as mise-v0) (#​570 by @​muzimuzhi).

New Contributors

Full Changelog: jdx/mise-action@v4.2.1...v4.2.2

v4.2.1: : Signed checksums and PATH export fix

Compare Source

A small patch release with two user-facing fixes: mise downloads are now verified against minisign-signed release checksums by default, and the env input no longer leaks the runner's PATH into subsequent steps.

Fixed

Verify mise downloads with signed checksums (#​548) by @​jdx

The action now embeds mise's minisign public key and verifies SHASUMS256.txt.minisig before trusting any release checksums, then checks the downloaded mise binary's SHA256 against the verified list. This applies to both GitHub release archives (verified before extraction) and the default mise.jdx.dev CDN path (verified against the signed checksum for the matching release asset). If a CDN download fails verification, the action warns and falls back to the signed GitHub release asset instead of installing an unverified binary.

  • The existing sha256 input still works as an explicit override.
  • Pinned mise versions older than 2024.12.24 (which predate minisign checksums) get a warning and skip signed verification rather than failing.
  • Because tar installs now extract from a verified file on disk, the previous streaming download | tar fast path is replaced with a download-then-verify-then-extract flow.

Thanks to @​potiuk for the detailed threat-model writeup in #​547.

Exclude PATH from environment export (#​556) by @​jdx

The env input has always documented that "PATH modifications are not part of this", but since the switch to mise env --json in #​252 (needed for redaction support), the action was exporting every string value returned by mise — including the computed PATH — into GITHUB_ENV. That effectively snapshotted the runner's entire PATH into subsequent steps and let [env] _.path entries in mise.toml leak past the action's own PATH management.

exportMiseEnv now skips PATH (case-insensitive) when exporting JSON env vars, restoring the documented behavior. Normal mise env vars are still exported, and PATH continues to be managed by the action's own setup (e.g. add_shims_to_path). Fixes #​555.

Full Changelog: jdx/mise-action@v4.2.0...v4.2.1

marocchino/sticky-pull-request-comment (marocchino/sticky-pull-request-comment)

v3.0.5

Compare Source

What's Changed

Full Changelog: marocchino/sticky-pull-request-comment@v3.0.4...v3.0.5

pypa/gh-action-pypi-publish (pypa/gh-action-pypi-publish)

v1.14.2

Compare Source

This one probably won't touch you visibly so just bookmark @​webknjaz's EuroPython 2026 “AI” slop rant for when it's published on YouTube or encourage him to come back with more to share next year!

🛠️ Urgh… Another release!? Again? Explain yourself!

Looking at the diff, you'll only witness updates across the dependency tree. That's it! It's not a security fix or anything like that even, no. But you'll want this update.

[!tip]
So what most people will find useful is @​takluyver💰's update of Twine to v7 that we use internally (#​416). This version will let them upload their sdists and wheels containing core packaging metadata v2.5 to (Test)PyPI.

🧐 Tell me why..

The rest of the updates bump things related to pypi-attestations and sigstore, which has the most interesting backstory here. @​facutuesca💰 sent a patch in #​417 but a bunch more helped out.

TL;DR non-pure-python projects with C-extensions tend to have dozens (sometimes hundreds) wheels to upload to PyPI per release. They are often quite big and take time to transfer over the network. People started noticing problems and coming up with DIY sharding workarounds like aio-libs/aiohttp#13226 around July 23.
On this date, projects with a good amount of bytes to publish would start getting timeouts 5 minutes after the PyPI publishing job begun. The same job that worked just fine before.

I had to start pinging upstream library and ecosystem people, on GitHub and privately, to start making sense of what was happening. Eventually, we collectively concluded that GitHub must've shortened the lifetime of their OIDC identity — it seems to have used to be 10 minutes long (at some point in the past) and is now 5 minutes, apparently. It's not documented clearly, and we have not been able to get any clarity by attempting to contact GitHub through private channels, using personal connections.

Over the course of investigation, @​facutuesca💰 found and fixed a related underlying cache invalidation bug in sigstore/sigstore-python#1838, which he then coordinated propagation through the dependency chain updates in sigstore-python, pypi-attestations, gh-action-pypi-publish and gh-action-sigstore-python.

Mike's also discovered that Sigstore's Rekor slowdown seems to have become the main contributing cause of the last week's incident. He's collected some data to support this claim: https://publishing-five-minute-timeout.tiiny.site.

Edge Cake XKCD feels just like this release

🫶 New Contributors

🪞 Full Diff: pypa/gh-action-pypi-publish@v1.14.1...v1.14.2

🧔‍♂️ Release Manager: @​webknjaz 🇺🇦

🙏 Special Thanks to @​davidbrochart💰 and @​Dreamsorcerer💰 for turning my attention (in #​415 and in private) to the newly surfaced corner case in GitHub's behavior that only affected a narrow category of projects while many others remained blissfully unaware. @​bdraco💰 came up with a DIY sharding workaround for aiohttp that served as a demo for other projects. @​miketheman💰 confirmed the Warehouse-side details. Also, @​jku💰 and @​woodruffw💰 helped work through, review and release the Sigstore ecosystem upstream libs.

💬 Discuss on Bluesky 🦋, on Mastodon 🐘 and on GitHub.

GH Sponsors badge

v1.14.1

Compare Source

This release was cut at EuroPython 2026 Sprints

🛠️ Internal Dependencies

@​adisivaprasad💰 helped get rid of the GitHub Actions runner warning about the old Node 20 runtime being used by updating actions/setup-python from v5.6.0 to v6.2.0 in #​408.

💪 New Contributors

🪞 Full Diff: pypa/gh-action-pypi-publish@v1.14.0...v1.14.1

🧔‍♂️ Release Manager: @​webknjaz 🇺🇦

🙏 Special Thanks to @​jylenhof💰 for reminding me to work on this release!

💬 Discuss on Bluesky 🦋, on Mastodon 🐘 and on GitHub.

GH Sponsors badge


Configuration

📅 Schedule: (in timezone Asia/Tokyo)

  • Branch creation
    • At any time (no schedule defined)
  • Automerge
    • At any time (no schedule defined)

🚦 Automerge: Enabled.

Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.

👻 Immortal: This PR will be recreated if closed unmerged. Get config help if that's undesired.


  • If you want to rebase/retry this PR, check this box

This PR has been generated by Mend Renovate.

@renovate-bot-lohn
renovate-bot-lohn Bot requested a review from lohn as a code owner July 12, 2026 14:44
@renovate-bot-lohn
renovate-bot-lohn Bot enabled auto-merge (squash) July 12, 2026 14:44
@renovate-bot-lohn
renovate-bot-lohn Bot force-pushed the renovate/github-actions branch 4 times, most recently from 1b1a5ab to b8928ba Compare July 23, 2026 15:29
@renovate-bot-lohn
renovate-bot-lohn Bot force-pushed the renovate/github-actions branch 4 times, most recently from adb9451 to 43c18af Compare August 1, 2026 21:06
@renovate-bot-lohn
renovate-bot-lohn Bot force-pushed the renovate/github-actions branch from 43c18af to 850ab92 Compare August 4, 2026 12:11
@renovate-bot-lohn
renovate-bot-lohn Bot force-pushed the renovate/github-actions branch from 850ab92 to b3e4573 Compare August 7, 2026 21:05
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants