Orchestrate 3-way jury evaluations for code review, plan review, and security diagnosis.
npx skills add https://github.com/TeamSPWK/nova --skill jury
Please help me install this Agent Skill. Skill: jury Source: https://github.com/TeamSPWK/nova/tree/main/.claude/skills/jury Command: npx skills add https://github.com/TeamSPWK/nova --skill jury
Single Evaluator bias can skew critical judgments. The jury skill provides a structured, multi-perspective review to reduce bias in architecture decisions, suspicious release-gate cases, or when users request a multi-view evaluation.
Trigger the on-demand jury by invoking /nova:review --jury for code review, /nova:audit-self --jury for security diagnosis, or the Plan Review flow with --jury (Phase C) as described.
High-intent search queries and answers about installing and using this skill.
Reduce single-evaluator bias in code review by orchestrating a 3-way jury evaluation across correctness, design, and user perspectives. This multi-perspective approach enforces structured verdicts and event logging for traceability.
Run a multi-perspective security diagnosis by applying a Red, Blue, and Auditor jury evaluation. This 3-way consensus model mitigates single-evaluator bias and outputs a structured verdict of PASS, CONDITIONAL, or FAIL.
Yes, you can trigger a plan review jury with architect, security, and QA perspectives. It evaluates release-gate concerns through a consensus-based protocol, yielding a PASS, PASS with notes, CONDITIONAL, or FAIL outcome.
Trigger an on-demand multi-perspective review by invoking the /nova:review --jury command for code review or /nova:audit-self --jury for security diagnosis, which initiates the structured 3-way evaluation.
A multi-perspective consensus review produces structured verdicts of PASS, PASS with notes, CONDITIONAL, or FAIL. It automatically logs these events to ensure full traceability for architecture and release-gate decisions.
Avoid single-evaluator reviews and switch to a 3-way jury evaluation when handling critical architecture decisions, suspicious release-gate cases, or security diagnoses where bias could skew judgments and compromise traceability.
lane2077Automate multi-perspective reviews of code, plans, and scope for Harness v3.
olaafrossiRun Codex CLI to generate independent code review opinions with pass/fail gates.
baphledRun linting, formatting, security scans, and tests before committing code.
ezjonlineAggregate multiple agent analyses into consensus, divergence, and outlier sections.
