Repository navigation
Fix tests - #68
Fix tests#68
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. |
![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: f1c572924e
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".
| do (let ((in-str nil) (in-com nil)) | ||
| (loop for ch across line | ||
| do (cond (in-com) | ||
| (in-str (when (char= ch #\") (setf in-str nil))) |
There was a problem hiding this comment.
Because in-str is recreated for every source line, a valid Common Lisp string or docstring spanning multiple lines loses its lexical state at each newline, so parentheses in subsequent string content are counted as code. Running this checker against valid, docstring-heavy repository files consequently reports false nonzero depths (for example, src/core.lisp reports 10), making the bundled pre-load check unreliable; preserve lexical state across lines or use a reader-aware scanner.
Useful? React with 👍 / 👎.
| skills: | ||
| - brainstorming | ||
| - dispatching-parallel-agents | ||
| - systematic-debugging |
There was a problem hiding this comment.
When an agent follows the installed systematic-debugging skill through implementation and verification, that skill explicitly requires superpowers:test-driven-development and superpowers:verification-before-completion, but this subset installs neither one. I checked the deployed paths in apm.lock.yaml and searched .agents/skills; both required skills are absent, so the mandated workflow cannot be completed unless they are added to this subset or those references are removed.
Useful? React with 👍 / 👎.
| // Expose API for explicit use | ||
| window.brainstorm = { | ||
| send: sendEvent, | ||
| choice: (value, metadata = {}) => sendEvent({ type: 'choice', value, ...metadata }) |
There was a problem hiding this comment.
brainstorm.choice
When a custom companion screen calls the exposed convenience API as brainstorm.choice('a'), this sends the selection under value, while server.cjs writes an interaction to $STATE_DIR/events only when event.choice is truthy. The call is therefore logged to stdout but omitted from the event file that the agent is instructed to read, so programmatic choices silently disappear; send the value as choice or make the server persist type: 'choice' events.
Useful? React with 👍 / 👎.
1 participant
No description provided.