Frontmatter
| id | 7687 |
| title | Enhance GitHub Workflow server robustness when gh cli is not installed |
| state | Closed |
| labels | enhancementai |
| assignees | MannXo |
| createdAt | Nov 1, 2025, 7:22 PM |
| updatedAt | Nov 10, 2025, 9:31 PM |
| githubUrl | https://github.com/neomjs/neo/issues/7687 |
| author | tobiu |
| commentsCount | 3 |
| parentIssue | null |
| subIssues | 7714 ci(testing): gate GitHub-integration tests and add diagnostics polish 7713 health(github-workflow): reproduce gh-absent |
| subIssuesCompleted | 2 |
| subIssuesTotal | 2 |
| blockedBy | [] |
| blocking | [] |
| closedAt | Nov 10, 2025, 9:31 PM |
Enhance GitHub Workflow server robustness when gh cli is not installed

Hi, As you mentioned, I have already partially addressed this in #7678 . I'm happy to work on this ticket as well, if you see fit.

this one definitely makes sense: uninstalling gh cli completely and see if the server handles it.
bigger picture: to get v11 ready, i would like to fully remove siesta. to do this, we need to explore which unit tests and component based tests did not get migrated yet, and tackle them. we also need to update some guides and the agents file itself (i am currently having sessions where gemini fails to use the memory core without several recovery prompts). you are definitely welcome to open new tickets for any topics of interest.
i am already moving new mcp testing files into the playwright folder.

closing the epic, since all subs are resolved.
The
HealthServicein thegithub-workflowMCP server currently checks forgh cliauthentication and version. While robust, the server's behavior and error handling whengh cliis entirely absent or not in the PATH needs further investigation. This ticket is to thoroughly inspect and enhance the server's startup robustness and error messaging in such scenarios.