Orchestrate parallel peer reviews of code and delivery artifacts.
npx skills add https://github.com/cryptopafi/nexusos-skills --skill multi-review-cryptopafi
Please help me install this Agent Skill. Skill: multi-review Source: https://github.com/cryptopafi/nexusos-skills/tree/main/multi-review Command: npx skills add https://github.com/cryptopafi/nexusos-skills --skill multi-review-cryptopafi
This Skill removes uncertainty from post-build deliveries by running multiple specialized reviewers in parallel to detect security issues, correctness bugs, and integration regressions before merge decisions.
Run /multi-review on the most recent NexusOS delivery to produce a PASS, CONDITIONAL, or FAIL gate and a detailed findings report.
High-intent search queries and answers about installing and using this skill.
Automated parallel peer review gates merges by spawning 3-5 specialized reviewer agents to evaluate security, correctness, integration, style, and performance concurrently. It aggregates verdicts under strict majority rules, blocking merges on any CRITICAL or security HIGH findings.
Post-build verification for code deliveries is best handled by orchestrating concurrent specialized reviewers that apply deterministic gating rules. This process evaluates integration regressions and correctness bugs, producing a PASS, CONDITIONAL, or FAIL gate alongside a detailed findings report.
Automated code review supports both standalone file-only mode and full workspace directory targets. You can apply the peer-review gate to individual files or entire NexusOS delivery directories for post-fix verification and CI pipeline gating.
Yes, the parallel peer-review gate blocks merges automatically on any CRITICAL findings and security HIGH findings. It enforces strict majority rules across all spawned reviewer agents and reports CONDITIONAL outcomes when applicable.
The code review process handles reviewer timeouts and skipped reviews through robust fallback mechanisms. It applies a 120-second soft timeout for each specialized reviewer agent and persists the aggregated review summaries to REVIEW.md and Cortex for trend analysis.
Using multiple specialized reviewers removes uncertainty from post-build deliveries by concurrently evaluating security, correctness, integration, style, and performance. This parallel approach detects integration regressions and correctness bugs more effectively than a single review pass before merge decisions.
randoneeringEvaluate Cortex Agents in Snowflake and review results in Snowsight.
MichielDeanPersist adversarial code review findings as self_assessment memories via the llmem track-review CLI.
fredmilhomeAutomate academic manuscript review with single-pass, adversarial, and peer-review modes.
sukeun1997Run parallel security reviews and fix critical findings with verification.
