chore: release v0.17.0 - #493
Conversation
|
Claude encountered an error after 1s —— View job I'll analyze this and get back to you. |
There was a problem hiding this comment.
Pull request overview
Bumps the project version to v0.17.0 and records the v0.17.0 release entry in the changelog, aligning the repository metadata with the release described in the PR.
Changes:
- Update
deva.shVERSIONfrom0.16.0to0.17.0. - Add a
## [0.17.0] - 2026-07-27section header toCHANGELOG.mdto start the v0.17.0 release entry.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
| deva.sh | Updates the script’s VERSION constant to 0.17.0. |
| CHANGELOG.md | Adds the v0.17.0 release header dated 2026-07-27. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 1fab342f78
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
|
|
||
| ## [Unreleased] | ||
|
|
||
| ## [0.17.0] - 2026-07-27 |
There was a problem hiding this comment.
Publishing the existing notes under the new 0.17.0 header makes them state that KIMI_CODE_VERSION is pinned at 0.28.0, while this commit's release tree has KIMI_CODE_VERSION=0.29.2 in versions.env. Anyone using the release notes to reproduce or audit the shipped image receives the wrong version, so update the Kimi pin stated in this release section to 0.29.2.
Useful? React with 👍 / 👎.
Minor bump: two new features, four fixes.
Added
others: kimi reads no API key from the shell, so api-key mode synthesizes
the KIMI_MODEL_* family in memory and mounts nothing.
-p cloakruns agents against CloakBrowser stealthChromium, headed on Xvfb, with VNC login, an always-on CDP-driven browser,
and a persistent host profile so a login survives container removal.
Image policy this release establishes
cloak is an auxiliary image on its own cadence, not a general releasable
profile: it holds no agent CLIs and only changes when its own inputs change.
So nightly skips it and the release cloak job is non-gating
(
continue-on-error, out of thereleasejob'sneeds). A ~200MB Chromiumbake per arch must not decide whether a release ships. #485 tracks turning
that policy into a profile registry instead of duplicated strings.
Fixed
environment and made
-pa silent no-op--dry-runpulled multi-GB images and rewrote the resolved tag.devamountThree of these are the same bash trap: a failing command substitution in a
plain assignment aborts under
set -e, quietly making the error handlerbelow it dead code.
Pins refreshed: claude-code 2.1.220, cctrace 0.23.0, gemini 0.52.0,
grok 0.2.112, kimi 0.29.2, ccx v0.12.0, playwright 1.62.0, cloakbrowser 0.5.2.
Tag goes on main's merge commit, not this branch head.