Repository navigation
Added linter and AI review to stack docs with intentional mistakes - #7
Conversation
Merged in an workflow YAML update that caused checks to fail.
AI documentation reviewdocs/stack.md:25 — “the the markdown files” contains a duplicated word. Change it to “Markdownlint checks the Markdown files for common formatting errors.” docs/stack.md:29 — “GPT 5.6 Luna” does not identify whether this is the model name, an internal alias, or a product component, so readers cannot determine what performs the review. Clarify the model/tool name or describe it as an internal alias. docs/stack.md:29 — “documentation drift” is undefined in this overview. Briefly explain what drift is checked for, such as inconsistencies between the changed documentation and the existing documentation. docs/stack.md:41 — The automation steps no longer mention checking out the repository or installing the tools, although the linter and site build require the source files and their dependencies. Restore those steps, or state that the workflow setup handles checkout and installation before these steps run. docs/stack.md:43 — “Run the AI review” does not explain how the review is invoked or what happens when it reports an issue. Identify the script or workflow step used and state whether findings cause the pull request check to fail or are only reported. |
AI documentation reviewdocs/index.md:40 — “Pull master” is ambiguous: it does not say which remote to pull from or how to create the branch from the updated base. Replace it with explicit commands such as docs/stack.md:41 — This says review occurs only “when a new pull request is created,” while the pipeline description says it also runs when commits are pushed to the PR branch. Change this to “when a pull request is created or updated” so contributors know when reviews are rerun. docs/stack.md:51 — The stack identifies GPT-5.6 Luna as the review model, but the setup instructions only provide the placeholder |
Purpose: test the AI review