Queue Arbiter Zebu decisions and track async human review status.
npx skills add https://github.com/CruxExperts/localsetup --skill ls-arbiter
Please help me install this Agent Skill. Skill: ls-arbiter Source: https://github.com/CruxExperts/localsetup/tree/main/_localsetup/skills/ls-arbiter Command: npx skills add https://github.com/CruxExperts/localsetup --skill ls-arbiter
Arbiter Zebu enables async human review of critical decisions to ensure architectural and planning choices are vetted before implementation.
Push a decision plan for human review by providing a title, a tag, and a set of decisions, then notify the assigned agent.
High-intent search queries and answers about installing and using this skill.
You can automate asynchronous human review by queuing decision plans with Arbiter Zebu and tracking their status locally. This ensures critical architectural and planning choices are vetted before implementation begins.
You need the arbiter-push CLI integrated into your environment to push plans for human review. The system maintains a local queue at ~/.arbiter to track pending plans until they are completed.
To push a batch decision plan, provide a title, a tag, and a set of decisions through the arbiter-push CLI. The plan is then queued locally and the assigned agent is notified for review.
Yes, you can monitor progress and fetch completed plans through the local queue maintained at ~/.arbiter. This allows you to track batch decision workflows and receive notifications when reviews are finished.
Asynchronous decision review is appropriate for plan design, architecture reviews, and batch decision workflows across software projects. It ensures critical choices are properly vetted before implementation proceeds.
Pushed plans remain pending in the local queue at ~/.arbiter until the human review process is completed. You can monitor progress and fetch completed plans once the assigned reviewer finalizes the decisions.
cptnfrenPush decision plans to Arbiter Zebu for asynchronous human review.
EpicenterHQWrite technical specifications with context, rationale, and open questions for implementation.
zhu637882-stackReview architecture, data flow, diagrams, edge cases, test coverage, and performance.
Geralt1983Facilitate AI-assisted pair programming with TDD, debugging, and code review.
