Update workflow-preset to v1.3.10 - #19
Conversation
bf16c01 to
2cf2117
Compare
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: bf16c0144b
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".
| - commit `tasks.md` only during `task_commit`; mark `[x]` only for receipt `completed_task_ids` that passed `receipt_review`, required code review, and `integration_verification` with no `deferred_validation_todos`; leave all other task checkboxes unchanged | ||
| - run `integration_verification`, closeout | ||
| - must not produce shard plans or digest drafts | ||
| - Follow the lifecycle and runtime isolation contract in `tests/contracts/speckit-cross-agent-subagents.md`. |
There was a problem hiding this comment.
When this preset is installed, the registered /speckit.implement prompt is copied as a standalone agent command and invoked from the consumer project root, but this new reference points the Core Agent at tests/contracts/speckit-cross-agent-subagents.md. In installed projects that file is not at that path (if present from the preset archive it lives under .specify/presets/workflow-preset/tests/contracts/...), so the shard/path/receipt rules that were removed from this command will not be loaded during normal /speckit.implement runs. Keep the runtime rules in the command or reference/package the actual installed path.
Useful? React with 👍 / 👎.
| @@ -343,8 +366,6 @@ def validate_design_requirement_intake_trace_contract(intake: dict[str, Any]) -> | |||
|
|
|||
| for row in rows: | |||
| item_id = row.get("visual_item_id", "<unknown>") | |||
There was a problem hiding this comment.
With the missing-ID guard removed, a single Visual Restoration Trace row without visual_item_id now passes validation because _duplicate_ids() only catches duplicate missing keys and the rest of the loop proceeds with "<unknown>". The template and README require one row per accepted Visual Item ID so those stable IDs carry traceability into spec.md and visual readiness; accepting rows without one lets malformed design intake reach later planning where evidence and tasks cannot be bound back to a visual item.
Useful? React with 👍 / 👎.
2cf2117 to
274df63
Compare
1 participant
Summary
Release artifact: https://github.com/bigsmartben/spec-kit-workflow-preset/releases/download/v1.3.10/spec-kit-workflow-preset-v1.3.10.zip
Test Plan
uv run --extra test python -m pytest tests/test_presets.py -k workflow_preset