Repository navigation
Update nix flake and plugin version for v0.15.0 - #930
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
There was a problem hiding this comment.
🟢 Approved
The three version-only changes are consistent with the intended release and have no identified blocking issues.
No open issues 🎉
What changed in this PR
Prepares v0.15.0 release metadata for the connector profile improvements in #907, following the release workflow used in #922.
Changes:
- Bumps the Nix package and both plugin manifests to 0.15.0.
- Leaves the Nix dependency hash unchanged.
[!TIP]
If you aren't ready for review, convert to a draft PR.
Click "Convert to draft" or rungh pr ready --undo.
Click "Ready for review" or rungh pr readyto reengage.
| File | Description |
|---|---|
| nix/package.nix | Sets package version to 0.15.0. |
| .codex-plugin/plugin.json | Sets Codex plugin version to 0.15.0. |
| .claude-plugin/plugin.json | Sets Claude Code plugin version to 0.15.0. |
Review effort: Balanced
…ecamp#791) * Add the basecamp subtasks umbrella on the SDK's Subtasks service basecamp/basecamp-sdk#883 models subtasks as a first-class resource on bc3's flat canonical routes. This surfaces all eight operations as `basecamp subtasks list|show|create|update|complete|uncomplete|move|delete`, account-scoped (no --in), taking a to-do or card as the parent. The SDK is pinned to basecamp-sdk main at 8216dd62c329 because basecamp#883 is not released yet; the pin must be swapped for the release that ships it before merge. The pin carries three SDK changes past v0.19.0 that the CLI has to absorb: - basecamp#930 split the Automation tag into Checkins, Templates, Webhooks, Lineup, Dock, Recordings and Search. The MCP automation domain claims them all, so its served surface is unchanged; Subtasks joins the todos domain. - basecamp#924 refuses malformed event-feed refusal bodies at the decode as a statusless malformed response. The MCP feed tests now assert that, and the filter-mismatch fixture uses digests of the documented shape. - basecamp#883 itself makes CardStepsService.Reposition refuse 0, since bc3 always counted from 1, so `cards step move --position` is now 1-based. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * subtasks: confirm the permanent delete, refuse parent URLs, reject negative --limit Review findings on basecamp#791: - `subtasks delete` is a permanent, non-trashable delete, so it now goes through ensureDeleteConfirmable and tui.ConfirmDangerous with --force, the same gate `chat delete` uses. - A to-do or card URL without a #__recording_ fragment names the parent; the per-subtask verbs no longer read its id as a subtask id. Subtask and card-step URLs, which name the subtask in the path, are accepted. - `subtasks list --limit -1` is a usage error, as on the other lists. - The skill's endpoint count follows API-COVERAGE to 203. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * subtasks: read a subtask id only from a URL that names one Codex and Copilot found two URL shapes subtaskIDArg still misread. One is a subtasks or card-steps collection URL, which carries the parent's id. The other is a numeric fragment on a URL that is not a to-do or card, which includes a bare #<id> and a fragment on a subtask URL. A fragment now counts only as #__recording_<id> on a to-do or card URL. A path id counts only on a single subtask or step URL that has no fragment. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * API coverage: the SDK pin is v0.21.0, which ships the Subtasks service * subtasks move: say --position must be 1 or more when it's given but invalid * cards step move: say --position must be 1 or more when it's given but invalid --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Rob Zolkos <rob@zolkos.com>
Release metadata for v0.15.0, which releases #907. The release script's own steps produced it (
scripts/update-nix-flake.sh 0.15.0,scripts/stamp-plugin-version.shandscripts/stamp-codex-plugin-version.sh):nix/package.nixversion 0.14.0 → 0.15.0.go.modandgo.sumare unchanged since v0.14.0, sovendorHashstays, and the script's Nix build succeeded with it..claude-plugin/plugin.jsonand.codex-plugin/plugin.jsonstamped 0.15.0.This goes in as a PR rather than the script's direct push to
main, as with #922. The v0.15.0 tag goes on this PR's commit: after it merges,make release VERSION=0.15.0from main finds the metadata in place and only creates and pushes the tag.Summary by cubic
Prepares release metadata for v0.15.0, which ships #907. Bumps the version in
nix/package.nixfrom 0.14.0 to 0.15.0 and stamps.claude-plugin/plugin.jsonand.codex-plugin/plugin.jsonwith 0.15.0.The
go.modandgo.sumare unchanged since v0.14.0, sovendorHashstays as-is. Merging this PR is what lets the v0.15.0 tag be created viamake release VERSION=0.15.0.Written for commit 85a9bba. Summary will update on new commits.